Jorge Valdivia is a technology and product leader who helps B2B SaaS companies scale across product, engineering, and platform strategy, with a focus on AI and enterprise value creation. As CTO at Fleetio, he operates at the intersection of product, technology, and business strategy, aligning architecture with revenue growth and long-term market positioning.AI has become the defining technology story of the last several years. With new models emerging almost weekly, companies race to add AI-powered features to their products, and businesses across industries seek ways to automate work and improve decision-making.Yet as AI becomes more accessible, access to AI is no longer a competitive advantage. The real advantage lies in the data, expertise, and operational context behind it. All software companies have access to the same foundational AI models, but what separates meaningful AI solutions from generic ones is the depth of information and quality of the product they're built on and their ability to apply that information to real-world customer problems.In fleet maintenance, where every decision affects costs, uptime, compliance, and safety, that distinction makes a difference. Customers need software that helps them make better decisions in the moments that matter. That's where the next generation of AI-powered service advisors is creating value.From Systems of Record to Systems of OutcomesFor much of the SaaS era, success was defined by becoming the system of record for a particular workflow. Businesses wanted a centralized place to store information, manage processes, and maintain visibility across their operations. That foundation remains important, but customer expectations have changed. Organizations are now asking what outcomes the software is helping achieve. The shift is subtle but significant. Customers no longer evaluate technology solely based on the information it collects; rather, they evaluate it based on the decisions it improves and the tangible results it delivers.This evolution is particularly evident in fleet management. Fleets generate enormous amounts of information, including repair records, inspection reports, telematics data, maintenance histories, vendor invoices, and diagnostic information. While the main challenge a few years ago was collecting data, it’s now become knowing what to do with it. The future of enterprise software will belong to platforms that can transform operational data into measurable business outcomes.Why an AI Service Advisor Needs More Than AIThe promise of an AI Service Advisor is to help maintenance teams make faster, smarter decisions, but the effectiveness of those recommendations depends entirely on the information available to the system. An AI Service Advisor trained on a limited dataset may be able to summarize information or identify broad trends, but a service advisor informed by more than a decade of real-world maintenance decisions and outcomes can do something far more valuable. It can identify patterns, predict outcomes, and recommend actions based on how similar situations have played out thousands of times before. The difference is context. Fleet operations are filled with decisions that appear simple on the surface but carry meaningful financial consequences. Should a repair be approved? Is a service recommendation necessary? Is a recurring fault likely to become a larger issue? Which maintenance actions will reduce long-term costs? The answers are rarely found in a single data point. They emerge from years of accumulated operational knowledge. This is why the most effective AI solutions are built on data, workflows, and domain expertise that have been refined over time.The Foundation MattersFor many organizations, AI feels like the beginning of a new chapter. In reality, AI is often the result of years of foundational work. Over the past decade, Fleetio has built a platform that captures maintenance activity, repair decisions, service histories, vendor interactions, and operational workflows across thousands of fleets. That foundation was built through years of customer collaboration, product development, and industry learning, resulting in a deep understanding of what effective fleet management looks like. That distinction matters because fleet management is not an intuitive domain. Unlike consumer software categories, where most users already understand the problem space, fleet operations require specialized knowledge developed through experience. Understanding maintenance strategies, asset lifecycles, repair economics, and operational tradeoffs takes time. As organizations look to deploy AI, those years of accumulated expertise become a significant competitive advantage. Models can be replicated, but domain knowledge cannot. The companies that will lead the next phase of AI adoption are the ones that have spent years building the right foundation that intelligence depends on.Turning Data Into DecisionsThe true test of analytics is whether it influences decisions. While many software platforms can identify trends after the fact, fewer can help customers act in the moment. Consider a common maintenance scenario. An asset is in the shop for service, and a vendor recommends additional repairs or replacement parts. Historically, a fleet manager would need to manually review the recommendation, compare it against maintenance records, and determine whether the work is justified.An AI-powered service advisor can surface relevant historical information instantly. It can recognize patterns, identify similar repairs across the fleet, and evaluate whether the recommendation aligns with historical outcomes. The value is in helping customers make a better decision at the exact moment that decision needs to be made. This is where analytics becomes a competitive differentiator. Instead of simply telling customers what happened, intelligent systems help determine what should happen next. When multiplied across hundreds or thousands of assets, even small improvements in decision quality can produce significant operational and financial impact. For one of our customers, Auto-Chlor, this technology resulted in $280K+ of avoided, unnecessary repair spend and a reduction in outsourced maintenance costs of about 15% in the past year. That’s a 4x return on investment for them, and that’s the kind of tangible results customers are looking for.Building Trust Through IntelligenceIn operational environments, trust matters as much as intelligence. Fleet managers operate in a world where maintenance decisions affect safety, compliance, budgets, and uptime. They need confidence that recommendations are grounded in reliable information and aligned with their operational realities. This is why Fleetio's approach to AI focuses on bounded intelligence rather than unrestricted autonomy. AI agents operate within defined workflows, evaluating specific conditions and helping customers navigate repetitive, time-sensitive decisions. An agent might assess whether a repair falls within policy thresholds or identify recurring asset issues that deserve additional attention. These systems augment human expertise rather than replace it. Humans still provide the context that data alone can’t capture. An asset's operating environment, local weather conditions, attached equipment, staffing constraints, or business priorities may influence a decision in ways that no model can fully anticipate, so the goal is for AI to help them focus their attention where it creates the most value.Analytics is the New Competitive MoatAs AI capabilities continue to mature, competitive advantages are shifting. The conversation is moving away from who has access to AI and toward who can generate the most meaningful outcomes with it. Increasingly, the answer comes down to analytics. Organizations that possess extensive proprietary datasets and deep domain expertise can deliver recommendations that are more accurate and more actionable than generic solutions. They can identify patterns others can’t see and provide guidance that reflects years of operational experience.For SaaS companies, this represents a significant shift. The most resilient businesses will help customers act on stored data by connecting historical records to present-day decisions and transforming workflows into intelligence, allowing AI to solve real problems rather than simply automate existing processes.The Future of Customer Problem-solvingThe future of enterprise software will be defined by who can best combine intelligence, expertise, and operational context to help customers achieve better outcomes. Platforms built on years of historical data and real-world operational knowledge possess a distinct advantage.Analytics is emerging as the true differentiator in the SaaS space. Organizations that succeed will be those that transform proprietary data into actionable intelligence, helping customers make smarter decisions with greater confidence.

Crypto networks normally operate as separate systems. Bitcoin, Ethereum, Solana and Layer 2 networks maintain their own ledgers and cannot automatically transfer assets or information between one another. Crypto bridges provide infrastructure that connects these isolated networks.Ethereum describes bridges as systems that allow tokens, information and even smart-contract instructions to move between blockchain ecosystems. How Does Crypto Bridging Work?When users bridge an asset, the original token usually does not physically “move” from one blockchain to another. Instead, bridge infrastructure uses mechanisms that represent its value on the destination network.Ethereum identifies three common approaches: lock-and-mint, burn-and-mint and atomic swaps. With lock-and-mint, for example, 1 ETH could be locked in a smart contract on the source network while an equivalent representation becomes available on the destination chain. When the user returns, that representation can be destroyed and the original asset released.This process differs depending on whether the destination uses canonical assets, natively issued tokens or externally bridged representations.Why Do Users Bridge Crypto?Cost and application availability are major reasons. A user might bridge ETH from Ethereum Mainnet to an Ethereum Layer 2 to access lower transaction costs. Others may move stablecoins to another blockchain to trade on a decentralized exchange, provide liquidity or use a lending application unavailable on their original network.The amount of capital involved is substantial. L2BEAT recently tracked approximately $33.4 billion in value secured across Ethereum scaling networks, including about $26.7 billion on rollups. Base accounted for roughly $10.98 billion and Arbitrum One around $10.12 billion. Native and Third-Party Bridges are DifferentA canonical bridge is generally associated with the blockchain or Layer 2 itself. Third-party bridges instead connect multiple networks through separate infrastructure.The distinction matters because every bridge introduces its own trust assumptions. Some depend heavily on smart contracts, while others rely on validators, multisignature systems or external messaging networks.Ethereum warns that bridges can expose users to smart-contract risks, technology risks and different levels of trust depending on their architecture.Bridges are becoming essential infrastructure as crypto develops into a multichain ecosystem. They allow users to move liquidity between networks and access applications that would otherwise remain isolated. However, bridging adds another layer of technical and security risk. Users therefore need to evaluate the bridge, asset representation, fees and destination network rather than simply assume that moving crypto across chains works like an ordinary wallet transfer.Why this Matters What Should Users Check Before Bridging?Users should first confirm the source network, destination network and exact token they will receive. Sending an asset through an unsupported route can lead to lost funds.Transaction fees also need attention because users may pay costs on the source chain, the bridge itself and eventually the destination network. A small test transaction can be useful before transferring a large amount.ConclusionCrypto bridges are becoming a critical part of the multichain ecosystem by allowing assets and liquidity to move between otherwise isolated networks. While they can unlock lower fees and broader DeFi access, users also take on additional smart-contract, network and trust risks. Choosing a reputable bridge and verifying every transfer detail remain essential.Also Read: Ethereum’s Next Phase: How the Blockchain is Preparing for the Next 10 YearsFAQs:1. What is a crypto bridge?A crypto bridge is infrastructure that connects separate blockchain networks and allows assets or information to move between them. It helps users access applications and liquidity that are not available on their original chain.2. Does crypto actually move from one blockchain to another?Usually, the original asset does not physically move between chains. Depending on the bridge model, tokens may be locked, burned or represented by an equivalent asset on the destination network.3. Why do users bridge crypto assets?Users often bridge assets to access lower fees, DeFi applications, decentralized exchanges or liquidity on another network. Ethereum users, for example, may move assets to Layer 2 networks to reduce transaction costs.4. What is the difference between canonical and third-party bridges?Canonical bridges are typically associated with the blockchain or Layer 2 itself, while third-party bridges connect multiple ecosystems through independent infrastructure. Each model carries different trust, validator and smart-contract assumptions.5. What risks should users consider before bridging crypto?Users should consider smart-contract vulnerabilities, incorrect networks, token representations, bridge fees and validator or multisig risks. Verifying the destination asset and testing with a small transaction can reduce the chance of costly mistakes.

Ahmedabad: Growth brings new opportunities for businesses, but it also introduces greater complexity into everyday operations. Workforce management is one such area. As employee numbers increase, businesses have to manage larger volumes of attendance records, leave requests, holidays, employee information and payroll-related processes. What may initially be handled through simple or manual systems can become increasingly difficult to coordinate as an organisation grows.This is creating a stronger case for integrating attendance and payroll as part of a broader workforce management approach.Attendance and payroll are closely connected functions. Attendance records provide important information for payroll processing, making accuracy and coordination between the two particularly relevant for businesses. When these functions are managed independently, HR and administrative teams may have to move information between different processes, verify records, and undertake repetitive coordination before payroll can be completed.An integrated approach can help simplify this workflow by bringing related workforce processes closer together. Instead of treating attendance and payroll as separate administrative responsibilities, businesses can view them as connected stages within their overall workforce management system.The need for such an approach becomes more apparent as businesses scale. A smaller organization may be able to manage workforce administration with relatively straightforward processes. However, as employee numbers increase, the volume of attendance data and payroll information also grows. Managing this information manually or across disconnected processes can increase the workload for HR and administrative teams.Technology can provide businesses with an opportunity to create greater structure around these activities. By connecting attendance and payroll, organizations can work toward reducing repetitive administrative intervention and creating a smoother flow of workforce information.The larger objective is not simply to automate one HR function. It is to make workforce management easier to manage as a business grows. When connected processes are organised within a more integrated framework, HR teams can potentially spend less time coordinating routine information and focus more closely on other workforce requirements.This changing approach to workforce management also provides the context for the transition from Petpooja Payroll to Attendo. What began as a solution designed to meet the workforce needs of restaurant businesses has evolved into a platform capable of addressing similar operational challenges across industries. The new identity reflects this broader applicability and positions the platform as a dedicated workforce management solution beyond its restaurant-focused origins.The transition also allows Attendo to move beyond being associated primarily with payroll. By combining attendance hardware with software capabilities, the platform is positioned around a more integrated approach to everyday people management.The shift reflects a broader evolution in how businesses are approaching HR technology. Rather than adopting technology for individual administrative activities in isolation, growing organizations are increasingly looking at how related functions can work together. This can be particularly relevant for processes such as attendance and payroll, where information from one function can directly affect the other.For growing businesses, this integration can become an important consideration when planning for scale. Workforce systems need to support an organization not only at its current size but also as employee numbers and operational complexity increase. Processes that depend heavily on manual coordination or separate systems can become more demanding over time, making simplicity and scalability increasingly important.This is particularly relevant for small and medium-sized businesses that may not have large HR teams or extensive administrative resources. Managing attendance hardware through one provider, payroll through another platform, and leave or employee information through additional tools can create unnecessary complexity. Bringing these functions into a connected system can make workforce administration easier to manage as the organization expands.The move toward integrated workforce management is therefore about more than convenience. It is about creating processes that are structured enough to support businesses through different stages of growth. When attendance and payroll are connected, organizations can approach employee administration through a more coordinated workflow rather than managing each activity independently.Attendo's new identity is positioned within this changing landscape. The transition from Petpooja Payroll gives the platform a broader workforce management narrative, allowing attendance and payroll to be communicated as interconnected elements of a larger HR technology proposition. The campaign itself is designed to move from the rebranding announcement toward product positioning and broader market authority.For growing businesses, the underlying requirement remains straightforward: workforce administration needs to become easier to manage as the organization grows in complexity. Integrated attendance and payroll can support that objective by bringing related processes closer together and creating a more organized approach to workforce information.As Indian businesses continue to adopt digital solutions for internal operations, connected workforce management is likely to become an increasingly relevant consideration. Integrating attendance and payroll is a step toward creating HR processes that are more structured, scalable, and aligned with the needs of growing organizations.

Key Takeaways :Batch API cuts eligible API costs by 50% while supporting large, latency-tolerant workloads.Batch queue limits depend on model, usage tier, and total queued input tokens.Tier 5 GPT-5-family models can reach a 15-billion-token Batch queue.OpenAI’s Batch API has become an important option for large AI workloads that do not need an instant result. The service lets businesses send a large set of API requests for later execution rather than ask for each result at once. OpenAI sets a 24-hour completion window for Batch jobs, while many jobs can finish sooner. The main financial benefit stands out clearly: Batch API requests cost 50% less than standard synchronous API use.This price cut can create major savings at scale. A workload that costs $1,000 through the normal API would cost about $500 through Batch, before any differences in model choice or input and output mix. At $10,000 of standard API spend, the nominal Batch cost falls to about $5,000. At $100,000, the same 50% reduction represents about $50,000 in savings.Batch Uses a Separate Capacity PoolBatch API does not rely on the same simple request-per-minute view that often guides standard API traffic. OpenAI measures Batch queue limits through the total number of input tokens held in the queue for a specific model. Tokens from pending batches count toward that limit. Once a batch finishes, those tokens leave the active queue.This detail matters for large AI workloads. A system with a high request-per-minute limit can still face a Batch queue limit if each request carries a large number of input tokens. Large document sets, long prompts and extensive datasets can therefore consume queue capacity much faster than a simple request count suggests.GPT-5 Family Shows the Scale DifferenceCurrent OpenAI model data shows how sharply capacity can rise across usage tiers. GPT-5.5 lists 500 requests per minute, 500,000 tokens per minute and a 1.5 million-token Batch queue at Tier 1. At Tier 2, the limits rise to 5,000 requests per minute, 1 million tokens per minute and a 3 million-token Batch queue.Tier 3 raises the Batch queue to 100 million tokens, while Tier 4 reaches 200 million. Tier 5 reaches 15 billion tokens, alongside 15,000 requests per minute and 40 million tokens per minute. The move from the 1.5 million-token Tier 1 queue to the 15 billion-token Tier 5 queue represents a 10,000-fold increase.GPT-5 and GPT-5.4 show the same broad Tier 1 through Tier 5 pattern in their current published limits. These figures make one point clear: there is no single OpenAI Batch limit. Capacity depends on the model and the account’s usage tier.Also Read - How AI Workloads are Changing Global Data CentersEmbeddings Can Reach Huge VolumesEmbedding workloads show an even wider capacity range. The current text-embedding-3-small limits list a 3 million-token Batch queue at Tier 1, 20 million at Tier 2, 100 million at Tier 3, 500 million at Tier 4 and 4 billion at Tier 5.That change takes the available Batch queue from 3 million to 4 billion tokens, or roughly 1,333 times more capacity. Such scale makes Batch well suited to large document sets, search indexes, semantic classification and offline data enrichment.24-Hour Window Changes Cost EquationThe 24-hour Batch window creates a simple trade-off. Standard API calls suit tasks that need an immediate result. Batch suits tasks that can wait. This distinction can make a large difference in total API cost.A company that needs results for a live customer request has little reason to select Batch. A company that needs to classify millions of records overnight has a very different requirement. The lower price and separate queue can make Batch a much better fit for that second case.Rate Limits Still MatterStandard API traffic still faces limits such as requests per minute and tokens per minute. A higher budget does not automatically remove those technical limits. Rate limits, spending limits and Batch queue capacity represent different controls.A large production system therefore needs a clear split between real-time and offline work. Live application requests can use the standard API, while large jobs can move to Batch. This separation can protect interactive traffic from large offline workloads and give each workload a more suitable capacity path.Batch Support Keeps ExpandingOpenAI has also widened Batch support during 2026. The API changelog shows Batch support for GPT Image 1.5, ChatGPT Image Latest, GPT Image 1 and GPT Image 1 Mini. OpenAI later added GPT Image 2 with Batch support and the same 50% Batch discount. Batch support has also expanded into video workloads, including Sora API use.This expansion shows that Batch no longer fits only text classification or embedding tasks. The same offline model now reaches more forms of AI work, including multimodal workloads.Reliability Still Needs AttentionLarge Batch systems still need careful job control. OpenAI Community reports from March and May 2026 described cases where some Batch jobs stayed at zero progress or remained in progress for long periods. These reports do not prove a general reliability problem across the service, yet they highlight a real operational concern for large deployments.A serious Batch setup needs status checks, expiration handling, retry plans and result reconciliation. A large dataset can contain thousands or millions of separate requests, so even a small failure rate can create a meaningful cleanup task.Also Read - OpenAI Realtime API: How It Works and When to Use It?The Key NumbersThe strongest figures tell the story clearly. Batch offers a 50% cost reduction and a 24-hour completion window. Several current GPT-5-family models list a 15 billion-token Batch queue at Tier 5, along with 15,000 requests per minute and 40 million tokens per minute. text-embedding-3-small reaches a 4 billion-token Batch queue at Tier 5.The central lesson is simple. Large API workloads should not rely only on request counts. Model choice, usage tier, input-token volume, queue capacity and required response time all shape the real throughput available. For workloads that can wait, OpenAI Batch API offers a powerful combination of lower cost and separate capacity. For real-time requests, standard API access remains the more suitable path.FAQsWhat is OpenAI Batch API?OpenAI Batch API handles large groups of API requests asynchronously within a 24-hour completion window.How much cheaper is Batch API?Batch API costs 50% less than standard synchronous API use for eligible workloads.What determines Batch API capacity?Model selection, usage tier, and the total input tokens in the active Batch queue determine capacity.What is the maximum GPT-5-family Batch queue?Several current GPT-5-family models list a 15-billion-token Batch queue at Tier 5.When should Batch API be used?Batch API suits large workloads that can wait for results, such as classification, embeddings, evaluations, and large-scale data processing.Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Overview:Free Fire MAX redeem codes offer players exclusive rewards without spending diamonds.Players can unlock skins, outfits, emotes, loot crates and vouchers.Codes are time-limited and region-specific, so redeem them quickly.Garena Free Fire MAX players can claim free in-game rewards using redeem codes released by the game’s developers. Some examples of items that can be unlocked include weapon skins, character skins, loot cases, and more, all without paying anything.It is important to note that redeem codes may be valid for only a short period and may even be restricted to certain regions.Free Fire MAX Redeem Codes for August 19The following redeem codes are available for Free Fire MAX players on August 19, 2026:BR43FMAPYEZZUPQ7X5NMJ64VS9QK2L6VP3MRFFR4G3HM5YJN6KWMFJVMQQYGFZ5X1C7V9B2NFT4E9Y5U1I3OFP9O1I5U3Y2TFM6N1B8V3C4XFA3S7D5F1G9HFK3J9H5G1F7DFU1I5O3P7A9SPlayers should note that not every code may work for every account. Redeem codes can be region-specific and may stop working once their validity period ends.How to Redeem Free Fire MAX CodesPlayers can redeem the codes through Garena’s official redemption website. The process is straightforward:Visit the official Free Fire MAX redemption website.Log in using the account linked to the game. Available login options include Facebook, Twitter, and Google.Enter one of the redeem codes in the designated text box.Click on Confirm to submit the code.If the code is valid, a confirmation message will appear.After a successful redemption, the rewards can take up to 24 hours to reach the player’s in-game mail.Also Read: Free Fire MAX Redeem Codes for August 18: Grab Exclusive RewardsWhat Rewards Can Players Get?Depending on the code used, players in Free Fire MAX can receive different rewards. Such rewards include legendary weapon skins, premium character outfits, exclusive loot crates, diamond vouchers, emotes, and pet skins.The rewards offered by each code might differ, and not every code will provide the same rewards. Why Free Fire MAX Redeem Codes Get Expired FastRedeem codes tend to expire quickly, especially because they are available for limited periods. The developers of the game might choose to restrict the use of redeem codes by offering them on specific servers or regions, meaning a redeem code can work on one server but not on others.For these reasons, it is advised that players use the redeem codes while they are still valid. Following the Free Fire MAX social media channels on Instagram, Facebook, and YouTube will also keep players up to date on codes and events.As these codes expire very often, checking their validity is recommended.Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Pi AI is an Empathetic AI conversation chatbot whose goal is to generate conversations that are natural, empathic, and emotionally intelligent. Pi AI acts as a conversational partner who can help users in their personal needs as well as in learning, idea generation, creative writing, organization, and other forms of conversations. The application does not focus on being a productivity chatbot but instead focuses on creating meaningful conversations. Pi AI works with voice interaction, multilingual capability, and personalized responses.General InformationHere are some known facts about Pi AI:Founded in: 2022Platform Support:  Web, Android, iOSDownload: Click HereMain Features of Pi AIBelow are some important Pi AI Features.Provides empathetic, human-like conversations with emotional intelligence.Supports natural voice conversations with multiple voice options.Offers multilingual conversations including English, Spanish, French, German, Italian, and Portuguese.Personalizes responses over time based on user preferences and conversation history.Assists with brainstorming, storytelling, creative writing, learning, reminders, goal tracking, and task organization.Benefits of Pi AIBelow are some Pi AI Benefits.Completely free without paid subscriptions or hidden charges.Creates natural conversations that feel friendly and supportive.Helps explain complex topics in simple language for learners.Offers voice interaction for a more engaging AI experience.Suitable for emotional support, idea generation, and everyday conversations.Challenges of Pi AISome Pi AI Challenges users may experience while using the platform.Offers shorter answers than more sophisticated AI chatbot programs.Inadequate abilities in coding, research, and other technical jobs.It is not easy at all times to navigate through the chat.Context switch is sometimes not done properly.Does not have a lot of features of productivity like other competing AI chatbots.Subscription InformationThe Pi AI application comes at zero cost. This means that users will not be required to subscribe to any paid plan in order to have access to all features. At the moment, there are no premium versions, hidden costs, or even charges depending on how the application is used.Support OptionsHelp Center and Learning Resources: FAQs, onboarding guides, AI conversation examples, product updates, educational resources, and usage tips.Customer Support: Online support, user feedback, technical assistance, account help, and mobile app support.ConclusionPi AI is an easy-to-use AI chatbot which focuses on emotional intelligence, positive conversation and customized conversations. It is one of the best AI companions if you are looking for friendly conversations, help in learning, brain-storming and even emotional support. Although it might not be as useful as AI assistants in programming, researching and generating content, but its conversational ability, voice input and complete free service makes it a great everyday AI companion.

The future of semiconductor design is increasingly driven by chiplet architectures and the Universal Chiplet Interconnect Express (UCIe) standard, one of the key topics at the IEEE/JSAP Symposium on VLSI Technology and Circuits 2026. As the industry moves from building one large chip to combining multiple specialized chiplets in a single package, engineers are tackling new challenges in high-speed data transfer, power efficiency, and signal integrity. To explore how this shift is changing the way modern chips are designed, we spoke with Shreesha Hegde, a Senior Analog Design Engineer with extensive experience developing advanced semiconductor technologies for global leaders including Samsung and NXP Semiconductors. Throughout his career, he has developed several key technologies, including communication chips for the Samsung Galaxy Watch 7 and Galaxy Watch 8, UCIe intellectual property supporting next-generation chiplet architectures, and power-efficient NFC technologies for Ingenico payment terminals deployed worldwide. His work has helped advance high-speed connectivity, low-power analog and mixed-signal design, and semiconductor architectures used in millions of consumer electronics and payment devices. Mr. Hegde, one of the key discussions at the Symposium was the industry's growing shift toward chiplet architectures and UCIe. From your perspective, what is driving this transition?I believe the biggest driver is complexity. For many years, the industry improved performance by making transistors smaller and fitting more functions onto a single chip. Today, AI, high-speed connectivity, and advanced consumer devices demand far more computing power, while users also expect longer battery life and faster performance. Building a single massive chip for every application is becoming less practical as designers balance performance, power, cost, and development time.Chiplet architectures give engineers much more flexibility. Instead of redesigning an entire chip, we can combine specialized chiplets and optimize each one for a specific function. In my view, that approach helps companies innovate faster and respond more quickly to changing market needs. That's why I see chiplets and standards like UCIe becoming an important part of the next generation of semiconductor design.The original communication chip you worked on for the Samsung Galaxy Watch 7 and Galaxy Watch 8 brought better connectivity and improved power efficiency to millions of devices. As semiconductor design becomes increasingly modular, what new engineering challenges does this shift create? In my experience, the biggest challenge is integration. While working on the communication chip for the Samsung Galaxy Watch, I saw how even small improvements in one part of the design could affect the performance and power efficiency of the entire system. Chiplet architectures create even more opportunities for optimization because engineers must design multiple specialized chiplets to operate as a single, efficient system.For me, the key question is always how to improve performance without increasing power consumption. Whether it's Wi-Fi, Bluetooth, GPS, or a future chiplet-based system, users expect fast connectivity, smooth performance, and long battery life. Meeting those expectations requires engineers to think about the entire system, not just individual components.Many experts describe UCIe as a technology that could do for chiplets what USB did for consumer electronics. You have developed UCIe intellectual property at Samsung to support next-generation chiplet architectures, placing you among the engineers helping advance this emerging standard. Based on that experience, do you agree with that comparison? Yes, I think it's a fair comparison. USB gave the industry a common way to connect devices, making technology easier to build and use. I see UCIe playing a similar role for chiplets. It gives semiconductor companies a common language to connect different chiplets, instead of reinventing the interface for every new design.Earlier at NXP Semiconductors, you developed power-efficient circuits for NFC chips used in Ingenico payment terminals worldwide. Today, AI hardware places even greater demands on power efficiency. How is this changing the role of analog and mixed-signal engineers?AI is changing the role of analog and mixed-signal engineers because power efficiency has become a system-level challenge rather than a circuit-level one. Early in my career, I learned that every milliwatt counts in chip design, because every efficiency improvement creates new opportunities for performance, functionality, and battery life. AI hardware raises the same challenge to a new level because every efficiency improvement creates more room for computing performance gains. That is why analog and mixed-signal design has become a bigger part of the overall architecture. Engineers now think beyond individual circuits and focus on how power delivery, signal quality, and high-speed communication work together to support the entire system.Having evaluated master's students in VLSI design and verification, what skills do you think future semiconductor engineers need most?That is a good question. Studying Electronics and Communication Engineering for my bachelor's degree and later earning a master's degree in Electrical Engineering gave me two different perspectives. My bachelor's program laid the foundation for my work in circuits, while graduate studies helped me think more deeply about system architecture and complex engineering problems. Serving as an evaluator for master's students in VLSI design and verification also gave me a chance to see how the next generation approaches these challenges. The students who stood out combined strong fundamentals with curiosity and a willingness to keep learning. Semiconductor technology changes quickly, so future engineers need more than technical knowledge. They need to understand how individual circuits fit into larger systems, adapt to new technologies, and keep asking the right questions.Finally, if we're having this conversation again in ten years, what do you think will have changed the most in semiconductor design? I think the discussions at this year's IEEE/JSAP Symposium gave us a good picture of where the industry is heading. Over the next decade, chiplet architectures and UCIe will become a natural part of semiconductor design, while AI will help engineers solve increasingly complex problems.For me, the most exciting part is that innovation will come from collaboration – between different technologies, different engineering disciplines, and people with new ideas. I believe that combination will define the next chapter of semiconductor design.

India’s Insolvency and Bankruptcy Code has improved recoveries in corporate insolvency cases, but personal guarantees remain a weak link in the resolution framework.According to the latest data from the Insolvency and Bankruptcy Board of India (IBBI), creditors have recovered only around 1% of admitted claims in cases involving personal guarantors.Since December 2019, creditors and companies have filed 5,186 applications seeking action against personal guarantors. However, resolution professionals have been appointed in only 2,137 cases, or roughly 41% of the total. This includes 51 appointments made through debt recovery tribunals.Only 64 Repayment Plans ApprovedThe gap becomes even wider at the resolution stage. So far, just 64 cases have resulted in approved repayment plans. Creditors have realized approximately Rs. 235 crore, translating into an average recovery of around Rs. 3.7 crore per case.This compares poorly with corporate insolvency cases, where recoveries have been around 31% of admitted claims.Personal guarantees are often provided by promoters to secure corporate loans or restructuring arrangements. If a company defaults and creditors fail to recover the full amount through the corporate insolvency process, lenders can invoke the guarantee against the promoter.Enforcement Remains DifficultThe low recovery rate highlights the difficulty of converting a personal guarantee into actual cash.Cases can face lengthy admission delays, while identifying and valuing a guarantor’s assets can be complicated. Assets may also be held through multiple entities or across jurisdictions, increasing the time required for recovery.According to the IBBI data, repayment plans are intended to establish a structured schedule under which guarantors repay creditors. However, the limited number of approved plans suggests that the mechanism is still moving slowly.Avoidance Transactions Add Another LayerIBBI data also shows that resolution professionals have identified avoidance transactions worth more than Rs. 4.6 lakh crore across 2,132 cases. These include alleged diversion of assets or other transactions that may have reduced the assets available for creditors.However, the regulator has not disclosed the amount actually recovered from these avoidance proceedings.Also Read: Supreme Court Sets Aside Essel Infra projects Insolvency Orders Over AI-Generated Fake Citations IBC Still Acts as a DeterrentDespite the weak performance of personal guarantee recoveries, the IBC has had a broader deterrent effect.More than 30,000 cases filed before the National Company Law Tribunal were reportedly resolved or withdrawn before admission, involving claims estimated at nearly Rs. 14 lakh crore.For banks and other lenders, the next major test will be whether faster admissions, better asset tracing and further legal clarity can improve recovery rates. Until then, personal guarantees remain a much weaker recovery tool than creditors may have expected when the framework was introduced.Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

At this point, artificial intelligence has proven it can answer questions. But nowadays, AI is starting to appear more and more in the e-commerce realm as we shift from conversational AI to action-focused, or agentic, AI. With standardized workflows and API-connected platforms, online retail is one of the first industries where AI is moving beyond recommendations to measurable operational outcomes.Data from StoreClaw, an autonomous commerce engine, comparing July 7-21 with the same period in June, found that seller interest in AI-powered product selection increased from 142 sellers (6.9%) to 197 sellers (11.5%). Similarly, interest in SEO and GEO optimization increased from 110 sellers (5.4%) to 131 sellers (7.6%).These figures reflect user intent signals rather than completed actions, but the direction is notable. Sellers appear to be using AI earlier in the planning cycle—when selecting products, preparing inventory, and improving discoverability ahead of seasonal demand.  This suggests that AI is becoming part of operational planning and not simply a productivity tool.¹Unlike consumer surveys, which measure purchasing intentions after they form, seller behavior can serve as one early signal of changing market activity. As merchants begin selecting products, preparing listings, and optimizing search visibility ahead of seasonal demand, AI platforms can provide an early view of how commerce is evolving.Beyond the AI Tool StackMost online sellers have already embraced AI. The challenge is that many have embraced multiple AI tools that operate independently from one another.According to reporting by Chinese technology publication 36Kr, cross-border merchants typically rely on more than 3.5 AI applications, while sellers operating independent stores often use five or more. ChatGPT generates copy, image generators like Midjourney create visuals, SEO tools optimize listings, and analytics platforms provide reports. Individually, these tools can save time. Together, however, they can also create fragmented workflows in which merchants still spend hours transferring information between systems.AI can accelerate individual tasks, but that does not necessarily mean the wider business process becomes more efficient.That is why the next stage of adoption is increasingly focused on moving from isolated AI assistance toward integrated AI execution. Sellers need systems that can connect workflows across multiple parts of the business rather than simply generate individual outputs.Point solutions can create content, but they cannot execute cross-platform business processes. Execution requires integrations, permissions, operational context, and closed-loop workflows that extend far beyond a single prompt.From AI Tools to AI ExecutionUnlike fragmented point solutions, StoreClaw is designed as a cross-platform commerce system that connects workflows across the channels where sellers already do business.At the core of the platform are more than 30 pre-built AI Skills:Packaged workflows covering product selectionSEO and GEO optimizationListing creationPricing analysisAdvertising optimizationCompetitive monitoringOther day-to-day e-commerce tasksThis approach also addresses a familiar problem with conversational AI: the blank chat window. Users do not necessarily need to know how to craft the perfect prompt before getting started. Instead, they can activate pre-built workflows designed around common e-commerce tasks.These workflows are supported by integrations with Shopify, Amazon, TikTok Shop, Instagram, and more than 20 other commerce platforms.Rather than relying only on generic examples or industry averages, StoreClaw connects directly to sellers' authorized business data—their actual products, inventory, sales, and marketing performance.Every recommendation is grounded in this business context, allowing analysis to reflect the seller's own operating conditions rather than a generic benchmark.Importantly, access to data does not mean handing over control of key business decisions. Execution follows only when the user approves it.The platform also extends beyond one-off interactions. Scheduled diagnostics, automated monitoring, and recurring optimization tasks mean that routine operational work can run continuously. Opportunities surface continually, as to any issues without constant intervention or oversight.The distinction is important. Point solutions can generate content or provide advice, but connecting the multiple systems involved in running an online business requires cross-platform data, permissions, and coordinated workflows.By packaging operational expertise into reusable AI Skills, StoreClaw aims to help sellers apply repeatable processes across multiple marketplaces while keeping people in control of important business decisions:Calculation runs are transparent and traceable.Publishing, price changes, and advertising edits require user confirmation.Listings can be reviewed for potential compliance issues before publishingThe Business Case for Agentic AIFor sellers, the practical value of agentic AI is better assessed through its tangible impact on business performance than through technical complexity alone.Public customer case studies published by StoreClaw illustrate how workflow automation translated into operational improvements across different types of e-commerce businesses.RuvalinoA Shopify maternity and baby products brand, Ruvalino consolidated six operational workflows into a single dashboard. According to the case study, repeat purchases increased from 11% to 18%,  while growing organic search share from 8% to 19%.INCENZOThis natural fragrance company reported 142% growth in organic traffic, a 3.4-fold increase in repeat subscribers, and automation of approximately 18 hours of manual SEO work each week.Twinkle StarLED decor seller Twinkle Star reduced new product launch time from five to seven days down to roughly one and a half days. Listing conversion increased from 9.3% to 14.1%, while quarterly gross merchandise value (GMV) grew by 120%.LuxClubAmazon home textiles brand LuxClub reduced advertising cost of sales (ACoS) from 35% to 22%, saving approximately $80,000 per month in advertising spend while increasing sales 47% quarter over quarter.Across these case studies, StoreClaw reports improvements in areas including traffic, revenue, advertising efficiency, customer retention, and operating costs.Its broader platform data points in a similar direction. During the first half of July, seller engagement increased across product selection, SEO/GEO optimization, and content creation. While these figures represent intent signals rather than completed execution, they indicate that merchants may be starting seasonal preparation earlier and incorporating AI more deeply into operational planning.Moving Into the Next Stage of AI AdoptionThere is also a straightforward economic argument behind this shift.Recent U.S. salary data from ZipRecruiter shows that the average e-commerce operations specialist earns around $55,000 per year,³ while experienced online business operations managers can earn close to $90,000 annually.At $39.90 per month, StoreClaw Max represents a relatively small software expense compared with the cost of adding dedicated operational headcount.⁵ The comparison is not a direct replacement for human expertise, but it illustrates why businesses are increasingly exploring AI for repeatable analytical and operational work.And labor savings are only part of the equation.Agentic AI increasingly makes it possible to evaluate commercial AI using familiar business metrics: revenue growth, operating efficiency, advertising performance, customer retention, and time saved on repetitive work.For technology and e-commerce leaders, the question is therefore shifting from whether AI will influence commerce operations to how quickly execution-focused systems will gain ground alongside collections of disconnected AI tools.The transition from advice toward execution is already underway. Businesses adopting AI-driven commerce workflows are beginning to report measurable outcomes, suggesting that the next competitive advantage in AI may come not simply from generating better answers, but from turning those answers into coordinated business actions.

TikTok is developing a feature that could allow users to send money to each other through direct messages, Bloomberg reported. The proposed service would use TikTok Pay, the company’s payment platform already available in parts of Southeast Asia for TikTok Shop transactions.References to the feature were reportedly found in code inside the current US version of TikTok’s iPhone app. The code suggests users could send money within a chat and add a message to the payment. Recipients would reportedly be able to accept transfers by tapping a payment prompt.TikTok told Bloomberg that the feature is not being tested. That indicates the project is still in its early stages. The company has not said when the feature could launch or whether it will be released to users.A peer-to-peer payment service would expand TikTok’s role beyond social media and online shopping. It could also put the platform in direct competition with established services such as Venmo and Zelle in the US.Wider Financial AmbitionsTikTok has been expanding its financial services plans in other markets. Reuters reported in March that the company had applied to Brazil’s central bank for permission to operate as a fintech firm. The proposed activities included lending and payment services.The potential payment feature also fits into TikTok’s broader push towards a wider digital ecosystem. The platform has added services including search, TikTok Shop, local discovery tools, games and hotel bookings.TikTok is not alone among social platforms in exploring payments. X recently began rolling out X Money in the US, giving users another way to send money through the platform.For now, TikTok’s direct-message payment service is only a development project. There is no confirmed launch date or assurance that the feature will reach the public.Also Read: Razorpay: Simplifying Digital Payments for Modern BusinessesJoin our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Overview :Bitcoin trades near $64,296, up 0.21% through today's session. Ethereum holds above $1,911, gaining 0.95% over the past day. Hyperliquid leads the top 10 gainers, up 1.40% over 24 hours.Bitcoin traded near $64,296 today, holding steady after extending gains through the early session. Ethereum climbed past $1,911, adding fresh support to broader altcoin sentiment. Trading desks flagged renewed spot demand across major tokens heading into the FOMC minutes release.FOMC minutes due later today continue to shape crypto positioning into September. A White House meeting between crypto executives and President Trump added fresh attention to Washington policy. Traders now track both events closely for the next clear directional signal.Bitcoin Price TodayBitcoin traded at $64,296.33, up 0.21% over the past 24 hours based on live CoinMarketCap data. Its market capitalization stood at $1.29 trillion. Trading volume over 24 hours reached $17.47 billion. Bitcoin has gained 0.79% over the past seven days, holding within a narrow band.Here is today's Daily Quote from leading crypto market analysts on Bitcoin's near-term structure.Bitcoin Faces a Low-Volume Liquidity TrapPrateek Gupta, Head of Business at Mudrex, shared that Bitcoin briefly reclaimed $65,000 as risk sentiment improved. President Trump's comments on the Strait of Hormuz lifted broader markets, though the macro backdrop stays challenging.He further mentioned the 30-year US Treasury yield climbed to 5.34%, its highest since January 2007. CryptoQuant data suggests the move could be a low-volume liquidity trap built on short covering. Gupta placed resistance near $65,000, with support shifting up to $64,000.Buyers Defend $64K as Yields ClimbCoinSwitch's Markets Desk noted BTC briefly touched $65,000 for the first time since August 10. A rebound in US equities helped, even as US-Iran tensions around the Strait of Hormuz persisted.He further added that whales have accumulated near $2.64 billion worth of BTC over the past two months. This points to continued dip buying despite the elevated yield backdrop weighing on risk assets. CoinSwitch flagged $65,000 as the key level needed to sustain momentum.Fed Hold Odds Firm Up Ahead of MinutesVikram Subburaj, CEO of Giottus, shared that Bitcoin traded near $64,300, up close to 0.3%. A Reuters survey found 94 of 104 economists expect the Fed to hold rates in September.He further noted large-cap altcoins showed limited momentum, with Solana near $76.88 and TRON near $0.3327. Subburaj advised investors to avoid chasing short-term moves through staggered entries and limited leverage. Immediate resistance sits near $65,000 to $66,000, with $63,000 as key support.Also Read: Bitcoin Holds $64,000 as $57,000 Long Liquidation Risk BuildsRebound Tests Resistance Near $65,000Riya Sehgal, Research Analyst at Delta Exchange, said Bitcoin's rebound from $62,700 is now testing the $65,000 zone. Sellers remain active around $64,800 to $65,200, capping the recovery so far.She further explained short covering and improving spot ETF flows drove the move higher. Bitcoin has reclaimed its major four-hour moving averages, keeping short-term structure constructive above $63,800. Sehgal flagged the FOMC minutes as the next likely trigger for direction.Crypto Prices Today: Top 10 Coins at a GlanceThe broader market holds a firm, range-bound tone today. Here is how the top 10 coins by market capitalization stand, based on live CoinMarketCap data.Biggest Gainers: SOL, HYPE, XRPSolana led today's top 10 gainers, up 1.52% over the past 24 hours. Hyperliquid followed closely with a 1.40% advance on steady demand. XRP also posted a firm 0.45% rise through the session.Biggest Losers: NoneEvery coin in today's top 10 traded flat to higher over the past 24 hours. Stablecoins USDT and USDC held their peg near parity. No major decliners appeared among today's largest cryptocurrencies.Crypto News Today: Top Headlines Impacting PricesFOMC minutes and the White House crypto meeting stand as today's biggest catalysts. Regulatory shifts and fresh institutional deals add further weight to sentiment.SEC Issues Regulation Crypto Proposal After Delayed VoteThe SEC issued its "Regulation Crypto" proposal today, days after canceling a meeting meant to vote on it. The delay had left capital-raising clarity in limbo for smaller digital asset issuers.Analysts note the SEC continues shaping policy through rulemaking rather than legislation this year. This approach offers faster relief but weaker permanence, since future administrations could reverse agency guidance quickly.Goldman Sachs Moves to Acquire Bitcoin ETF Manager NEOSGoldman Sachs is buying bitcoin ETF manager NEOS in a deal valued at up to $2.25 billion. The acquisition lets Goldman expand its footprint in the competitive spot Bitcoin ETF space.Bloomberg analyst Eric Balchunas said the move could help Goldman leapfrog BlackRock's rival fund lineup. The deal signals traditional banks see durable long-term demand for regulated crypto investment products.Bitcoin ETF Inflows Return After Three-Day Outflow StreakUS spot Bitcoin ETFs recorded net inflows on Monday, snapping three consecutive sessions of outflows per Coinglass data. The rebound follows renewed institutional appetite as Bitcoin pushed back toward the $65,000 level.BlackRock's IBIT fund still carries the largest share of institutional capital among issuers tracked today. Ether funds saw milder flow activity by comparison, suggesting spot demand continues absorbing broader caution.FASB Proposes Treating Certain Stablecoins as Cash-LikeThe Financial Accounting Standards Board (FASB) proposed that certain stablecoins should qualify as cash-like instruments under accounting rules. The move could reshape how corporations report stablecoin holdings on balance sheets going forward.Industry watchers see the proposal as a step toward mainstream stablecoin adoption within traditional finance. Clearer accounting treatment may encourage more corporate treasuries to hold stablecoins as working capital.Cuomo Backs CLARITY Act as Key Link to Traditional MarketsFormer New York Governor Andrew Cuomo said the CLARITY Act remains key to linking crypto and traditional markets. His comments come as the bill faces continued delays ahead of the Senate calendar.The remarks add political weight to today's White House meeting between crypto executives and the administration. Traders continue pricing regulatory clarity as a meaningful catalyst for institutional capital allocation decisions.Kalshi Ordered to Geofence Washington Users by TodayA King County court ordered prediction market Kalshi to block Washington state users starting today, citing gambling law violations. The ruling followed a CFTC emergency intervention to protect Kalshi's trading operations days earlier.Washington's Attorney General accused the platform of running an unlicensed gambling operation through event contracts. The case adds to growing regulatory scrutiny facing prediction markets operating alongside crypto exchanges nationwide.Also Read: Crypto News Today: Bitcoin Inflows, Jane Street Disclosed BTC Holdings, Bitmine Adds 9,926 ETHInvestor and Market OutlookBitcoin holds near $64,296 as traders weigh firming Fed hold odds against fresh regulatory headlines. A close above $65,000 could open a path toward stronger resistance near $66,000. A slip below $63,800 risks exposing the $63,000 support region next.Ethereum's hold above $1,911 adds a constructive signal for broader altcoin sentiment this week. The White House crypto meeting and FASB's stablecoin proposal keep regulatory focus sharp. FOMC minutes and ETF flow continuity remain the two catalysts traders are watching most closely.FAQsWhat is the Bitcoin price today? Bitcoin trades near $64,296.33, up 0.21% over the past 24 hours through today's session. Resistance builds near $65,000 and $66,000, while support sits near $63,800, based on levels shared across major trading desks today.Why did Bitcoin move higher today? Steady spot demand and returning ETF inflows supported Bitcoin's hold above $64,000. Ethereum's move past $1,911 also lifted broader sentiment across altcoins. Traders remain cautious ahead of today's FOMC minutes release and its policy signals.What is the biggest crypto news today? The White House crypto meeting and the SEC's new Regulation Crypto proposal stand as today's dominant headlines. Goldman's NEOS acquisition, FASB's stablecoin proposal, and the Kalshi ruling add further important context.Which coins are performing best today? Solana leads today's top 10 gainers near 1.52%, followed closely by Hyperliquid and XRP. No coin among the top 10 posted a decline today, with stablecoins USDT and USDC holding steady near parity.What should investors watch this week? Track FOMC minutes, Bitcoin ETF flow continuity, and today's White House crypto meeting outcome closely. CLARITY Act developments and Treasury yield movement could also influence near-term positioning across the broader market.

It’s no surprise that data is incredibly important for trading platforms, just as it’s crucial with any online service. However, traders should know exactly how exchanges send out massive datasets, how brokers get these feeds from the market, and how they use them to accurately price securities and derivatives.It Starts With Exchange DataMarket data begins with the exchange, whether it’s the NYSE, the NASDAQ, or an overseas organization. Every trading day, they log every trade in their order books. This takes in everything, limit orders, canceled orders, modified orders, and completed orders. Each order covers the ticker, the bid/ask, the time, and other pertinent details. Naturally, this results in a lot of data.Every ticker generates hundreds to thousands of trades per second, all of which get logged. Experienced traders often refer to this log as “the tape” (a holdover from when it was literally printed on tape). There are also Bloomberg Terminals that show real-time market data to users who pay a high annual fee.How Platforms Get Data FeedsJust like Bloomberg Terminal subscribers, trading brokers pay a premium to get data feeds. They get it from market data vendors (including Bloomberg), who themselves use cleaned up data taken from exchanges or third-party consolidators. The Securities Information Processor does this in the US, providing a unified feed for all American exchanges.With the dawn of High-Frequency Trading (HFT) run by algorithms, getting a near-instant data feed is essential for a fintech platform and its clients, traders operating in a market where HFT occurs.Source: UnsplashMost platforms subscribe to these data feeds and plug them into their service. There, it interacts with the broker’s app/site and the charts they support. That’s enough for stock trading, where the spread is just the bid/ask, showing the highest or lowest prices the security is trading for. The gap between bid and ask will depend on the liquidity and volatility of the security.How Derivatives Decide on SpreadsDerivative trading is more decentralized. Forex and bonds trade over the counter (OTC) between dealers, which then get aggregated into useable data for trader-facing platforms. CFD contracts work differently, being a contract between a trader and a CFD broker that creates its own spreads. When CFDs speculate on underlying stocks, the broker still takes market data into account. They reference real security prices as logged by exchanges, then add a slight markup to the bid and the ask.This means that ultimately, the size of the spread and the markup depends on volatility and liquidity. Small caps often have wider spreads for this reason. CFDs on non-stock entities, like forex and commodities, use the aggregated vendor data taken from OTC trades instead.Modern exchanges wouldn’t be able to operate without the ability to capture massive amounts of data, repackage it, and redistribute it to platforms. Those platforms then allow retail traders to have their say in the market. Some CFD brokers add a small markup as the cost of doing business on a free platform.The information provided does not constitute investment research. The material has not been prepared in accordance with the legal requirements designed to promote the independence of investment research and as such is to be considered to be a marketing communication.All information has been prepared by ActivTrades (“AT”). The information does not contain a record of AT’s prices, or an offer of or solicitation for a transaction in any financial instrument. No representation or warranty is given as to the accuracy or completeness of this information.Any material provided does not have regard to the specific investment objective and financial situation of any person who may receive it. Past performance is not a reliable indicator of future performance. AT provides an execution-only service. Consequently, any person acting on the information provided does so at their own risk.

Gold traded lower on MCX on August 19 as elevated crude oil prices, amid stalled US-Iran talks, fueled US Fed interest rate hike bets. August gold futures fell 0.2% to Rs. 1,53,953, and September silver futures declined 1.41% to Rs. 2,29,143. Meanwhile, Brent crude futures rose 0.71% to $91.67 per barrel. US West Texas Intermediate (WTI) edged higher by 0.80% to $85.62 per barrel.The CME FedWatch Tool indicates markets are currently pricing in a 65% probability that the Fed will keep rates unchanged and a 35% chance of a rate hike in September.Domestic Gold Prices24K gold fell by Rs. 90 to Rs. 1,54,970 per 10 grams, while 22K gold also declined by Rs. 85 to Rs. 1,42,050. By city, Mumbai and Kolkata mirrored prices at Rs. 1,54,970, while Delhi was at Rs. 1,55,120, and Chennai at Rs. 1,54,970.US Gold PricesUS gold prices rose on Wednesday as Treasury yields eased, with investors ‌awaiting minutes of the US central bank's July meeting for fresh clues on the monetary policy outlook. Spot gold climbed 0.5% to $4,356.55 per ounce. ​US gold futures slipped 0.2% to $4,410.20.Spot ​silver slid ⁠0.4% to $63.03 per ounce, platinum gained 0.5% to $1,720.10 and palladium held steady at $1,289.45.Also Read: How Gold Can Strengthen Your Long-Term Financial Plan in 2026Key Levels to Watch“Reduced expectations for ​Federal Reserve interest rate hikes and rising fiscal budget concerns are positive factors for gold,” ​said Kelvin Wong, a senior market analyst at OANDA."A sustained break above $4,390 could open the door (for gold) towards $4,505, while a break below $4,300 could expose $4,200 and $4,150," said Lukman Otunuga, head of market research at FXTM.On MCX, gold has support at Rs. 1,53,100 and Rs. 1,52,200, and resistance at Rs. 1,55,000 and Rs.1,56,100, while silver has support at Rs. 2,30,000 and Rs. 2,26,600, and resistance at Rs. 2,23,500 and Rs. 2,37,700.Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Key Takeaways - Best value: realme P4 Pro combines a 7,000mAh battery, 80W charging, 144Hz AMOLED display and Snapdragon 7 Gen 4 for Rs. 28,999.Best performance: OnePlus Nord 5 is the fastest option, but a genuine sub-Rs. 30,000 deal is currently difficult to find.Best design: Motorola Edge 70 stands out with its ultra-slim 159g body, IP69 protection and wireless charging.The sub-Rs. 30,000 5G smartphone market has become highly competitive in India. OnePlus, Motorola and realme now offer phones with fast processors, high-refresh-rate AMOLED screens, large batteries and capable cameras at prices close to each other. Among the latest choices, the OnePlus Nord 5, Motorola Edge 70 and realme P4 Pro 5G stand out for different reasons.The realme P4 Pro 5G offers the strongest overall value at Rs. 28,999. The Motorola Edge 70 costs about Rs. 27,999 and focuses on a slim design, durability and a premium feel. The OnePlus Nord 5 remains the most powerful option, although its sub-Rs. 30,000 price has limited availability at present.OnePlus Nord 5: The Performance ChoiceThe OnePlus Nord 5 has the most powerful hardware among these three phones. Its Snapdragon 8s Gen 3 processor sits above the Snapdragon 7 Gen 4 chip found in both the Motorola Edge 70 and realme P4 Pro. Tests put the Nord 5 at about 1.48 million points in AnTuTu, which gives it a clear advantage for demanding games and heavy daily tasks.The phone has a 6.83-inch 1.5K AMOLED screen with a 144Hz refresh rate. A 6,800mAh battery supplies power, while 80W charging offers fast top-ups. The camera setup has a 50MP main sensor with OIS and an 8MP ultrawide camera. A 50MP front camera handles selfies and video calls, while the phone can record 4K video at 60fps.OnePlus also gives the Nord 5 four Android upgrades and six years of security updates. That support period adds strong long-term value.There is one major issue with the price. A Rs. 28,760 listing appears in current market data, but that listing is unavailable. OnePlus lists the Nord 5 from Rs. 33,999 on its own site. The sub-Rs. 30,000 price therefore works as a deal target rather than a dependable current price.Buy Nowrealme P4 Pro: The Best Overall ValueThe realme P4 Pro 5G offers the most balanced package at Rs. 28,999. The phone uses the Snapdragon 7 Gen 4 processor and pairs it with a large 7,000mAh battery. An 80W charger handles fast power recovery, while the 6.8-inch AMOLED panel offers a 144Hz refresh rate.The camera setup includes a 50MP Sony main sensor with OIS and an 8MP ultrawide camera. A 50MP front camera adds strong hardware for selfies and video calls. The phone also supports 4K video at 60fps and uses UFS 3.1 storage.The P4 Pro reaches about 1.09 million points on AnTuTu in independent tests. Battery tests also show strong endurance, with a PCMark result of about 17.3 hours. A full 20 to 100 percent charge takes about 53 minutes in one test.The phone also has dedicated graphics hardware for games. realme claims support for BGMI at up to 144fps with its dual-chip setup.Software support remains weaker than the Nord 5. The P4 Pro gets three OS upgrades and four years of security updates. Still, at Rs. 28,999, the hardware package makes the P4 Pro the strongest value choice for most buyers.Buy NowAlso Read - iPhone vs Pixel: Google Pixel Wins the AI Photo Editing TestMotorola Edge 70: The Premium Design PickThe Motorola Edge 70 takes a different approach. Its Snapdragon 7 Gen 4 processor matches the realme P4 Pro, but its main focus lies elsewhere.The phone has a 6.7-inch 1.5K AMOLED display with a 120Hz refresh rate. Its 4,800mAh silicon-carbon battery supports 68W wired charging and 15W wireless charging. The camera system has two 50MP rear sensors, while the front camera also uses a 50MP sensor.The Edge 70 stands out with a 5.99mm body and a weight of only 159 grams. IP69 protection adds strong resistance against dust and water. Android 16 comes preinstalled.The smaller battery gives the realme P4 Pro an advantage for endurance, while the Snapdragon 7 Gen 4 keeps both phones close in everyday performance. The Edge 70 makes more sense for buyers who value a thin body, low weight, wireless charging and strong durability.Buy NowWhat About the Cheaper OnePlus Nord CE 6 Lite?OnePlus also offers the Nord CE 6 Lite below the main Rs. 30,000 options. Prices start at Rs. 20,999 for 6GB/128GB, Rs. 22,999 for 8GB/128GB and Rs. 25,999 for 8GB/256GB. The phone uses the Dimensity 7400 Apex chip, a 7,000mAh battery and 45W charging.The regular Nord CE 6 now sits above the Rs. 30,000 limit in current retail listings, with prices around Rs. 34,829 to Rs. 34,999. It therefore does not qualify as a true sub-Rs. 30,000 choice at present.Also Read - 7 Best MagSafe Accessories for Android Phones Worth Buying in 2026Final VerdictThe realme P4 Pro 5G takes the best-value position at Rs. 28,999. Its 7,000mAh battery, 80W charging, 144Hz AMOLED screen, capable cameras and Snapdragon 7 Gen 4 chip create a strong package for the price.The OnePlus Nord 5 remains the clear performance winner. A genuine in-stock price below Rs. 30,000 would make it the strongest choice for gaming, speed and long-term software support.The Motorola Edge 70 wins for design, comfort, durability and wireless charging. At around Rs. 27,999, it offers a different type of value.For an actual sub-Rs. 30,000 purchase today, the realme P4 Pro 5G stands as the safest overall recommendation. A genuine Nord 5 deal at Rs. 29,999 or less would change that verdict in favour of OnePlus.FAQs1. Which is the best phone under Rs.30,000?The realme P4 Pro 5G is the safest overall recommendation at Rs.28,999 as it balances performance, battery life, display quality, cameras and charging.2. Is the OnePlus Nord 5 better than the realme P4 Pro?Yes, for performance. Its Snapdragon 8s Gen 3 is substantially more powerful, making it the better choice for demanding gaming and heavy workloads.3. Which phone has the best battery life?The realme P4 Pro has the largest battery at 7,000mAh, giving it an advantage over the Nord 5's 6,800mAh and Edge 70's 4,800mAh batteries.4. Why choose the Motorola Edge 70?Choose it for its 5.99mm slim body, 159g weight, IP69 protection, 120Hz AMOLED display and 15W wireless charging.5. Should you buy the OnePlus Nord 5 under Rs.30,000?Absolutely, if you find a genuine in-stock deal at Rs.29,999 or less. Its stronger processor and longer software support would make it the top performance choice.Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Key Takeaways -SraVaani-1.0 covers 65 Indian languages and dialects, with the released checkpoint listing 63.The model shows strong results across low-resource, tribal and dialect languages.SraVaani is open source and available through Hugging Face for further development.India has taken a major step in speech technology with SraVaani-1.0, an open-source automatic speech recognition model from SPIRE Lab at IISc and ARTPARK@IISc. The new system can handle 65 Indian languages and dialects, with a strong focus on languages that receive little support from major speech AI systems. The SraVaani-1.0 research paper appeared on arXiv on August 8, 2026.The project stands out for its focus on India's long list of low-resource languages. Large commercial speech systems often perform best on languages with huge amounts of digital speech data. Many Indian languages and local dialects do not have that advantage. SraVaani takes a different route. Its creators built the system around a large Indian speech corpus and added methods that help the model learn useful links between speech, language and visual information.Huge Speech BaseSraVaani starts with the VAANI corpus, which contains a much wider range of Indian languages. The first stage used 31,255 hours of unlabelled speech for self-supervised speech pretraining. VAANI itself covers 105 languages across its broader corpus. This large base gives the model exposure to speech patterns that standard datasets often miss.The next stage adds an unusual feature. SraVaani uses about 11.85 million paired audio-image samples to align speech with visual information. This approach helps the speech encoder connect spoken words with meaning and context. The idea matters for languages with limited labelled speech, where extra forms of useful information can help improve recognition.The final stage uses about 31,263 hours of labelled Indian speech from 24 public datasets. This stage covers the reported 65 languages and dialects. Together, these stages give SraVaani a broad base for speech recognition across India's diverse language landscape.65 Languages, With an Important DetailThe research paper reports coverage across 65 Indian languages and dialects. The current Hugging Face model card, however, describes the released SraVaani-1.0 checkpoint as supporting 63 languages and dialects. This small difference matters when the project gets described in technical reports. The safest description remains that the SraVaani-1.0 research system covers 65 languages and dialects, while the current released checkpoint lists 63.SraVaani uses the FastConformer architecture with a hybrid TDT-CTC decoder. The released model has about 430 million parameters and takes roughly 900 MB in FP16. These figures place the system within a practical range for developers who need a capable multilingual speech model rather than a very large general AI model.Also Read - Top Multilingual Text-to-Speech Tools in 2026Strong Results for Low-Resource LanguagesThe most important result does not come from the language count alone. SraVaani shows particular strength in languages that lack strong speech AI support.A recent ARTPARK evaluation compared SraVaani with Google Gemini 3 Flash, Sarvam Saaras v3 and IndicConformer-600M-Multilingual across eight benchmarks. SraVaani recorded the lowest word error rate on many language-dataset combinations. The results also showed strong performance on several low-resource languages and dialects.Of 17 languages where a direct comparison with IndicConformer proved possible, SraVaani recorded a mean word error rate of 28.4%, compared with 30.2% for IndicConformer. A lower word error rate means fewer mistakes in the final transcript.The results become more notable for tribal and dialect languages. ARTPARK reports that SraVaani produced output for 51 tribal and dialect languages in the relevant VAANI benchmark region where competing systems either produced no output or recorded word error rates above 80%.An Open Model for DevelopersSraVaani has also moved beyond a research paper. SraVaani-1.0 is available on Hugging Face, and its model repository lists an MIT license. Access to the model files currently requires acceptance of Hugging Face conditions. ARTPARK also provides a live SraVaani demo and continues to update its model collection.Recent project activity shows a fast development cycle. The technical paper arrived on August 8, followed by an ARTPARK article on August 10 that explained the role of vision and audio alignment. On August 11, ARTPARK published a guide for fine-tuning SraVaani with custom speech data. The organisation's Hugging Face page also shows recent updates to SraVaani-1.0 and a SraVaani-0.5-live model for real-time speech-to-text use.Also Read - Top AI Voice Generator and Text-to-Speech PlatformsWhy SraVaani Matters?SraVaani's real value lies in its reach. India's speech technology cannot serve the whole country if strong performance remains limited to a small group of major languages. A system that can recognise tribal languages, regional dialects and other low-resource forms of speech can expand access to voice-based technology.The model could support local-language government services, education tools, accessibility products, healthcare interfaces and voice assistants. Its open availability also gives researchers and developers a foundation for new applications and further language work.SraVaani therefore represents more than a 65-language milestone. It shows a clear attempt to build speech AI around India's full linguistic diversity, rather than only its largest digital languages. The latest results suggest that this approach can produce useful gains precisely where mainstream speech systems often struggle most.FAQs1. What is SraVaani?SraVaani is an open-source automatic speech recognition model developed by SPIRE Lab at IISc and ARTPARK@IISc.2. How many Indian languages does SraVaani support?The research paper reports coverage across 65 Indian languages and dialects, while the current released checkpoint lists 63.3. What makes SraVaani different from other speech models?SraVaani places strong emphasis on low-resource, tribal and dialect languages that often have limited support from mainstream speech systems.4. Is SraVaani available to developers?Yes. SraVaani-1.0 is available on Hugging Face under an MIT license, with access to model files subject to Hugging Face conditions.5. How accurate is SraVaani?Across 17 languages with a direct comparison, SraVaani recorded a mean word error rate of 28.4%, compared with 30.2% for IndicConformer.Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Tesla plans to introduce its Cybercab robotaxi service in Austin, Texas, before the end of August. The company will reportedly offer the first public-road rides to employees. Cybercabs could join Tesla’s existing Austin robotaxi fleet several days later. People familiar with the preparations said the launch schedule ‘could still change.’Tesla Cybercab Launch Starts with StaffTesla has informed employees about preparations for the initial Cybercab rollout, according to The Information. The company plans to carry staff on public roads before opening wider access.These controlled rides will allow Tesla to monitor the vehicles under normal traffic conditions. The company has already carried employees along private routes near its Austin facility.Moreover, Tesla recently trained local police, fire, and rescue teams. The sessions covered Cybercab operations and possible emergency responses involving vehicles without standard driving controls.Tesla has not publicly confirmed the reported August launch schedule. It also did not immediately respond to media requests for further information about the rollout.The company followed a similar testing process before opening its Austin robotaxi operation. Tesla initially limited access while monitoring vehicle performance inside a defined service area.Public-Road Tests Support Austin RolloutTesla started testing the production version of Cybercab on public roads in June. These tests followed earlier development work at the company’s Texas manufacturing site.Chief Financial Officer Vaibhav Taneja confirmed factory testing during Tesla’s second-quarter earnings call. “We’ve started doing Cybercab rides in our factory in Austin,” Taneja said.The Cybercab carries two passengers and uses Tesla’s autonomous driving technology. Unlike the company’s current passenger vehicles, it has no steering wheel or pedals.Tesla designed the vehicle specifically for driverless transport rather than private ownership. Even so, board chair Robyn Denholm previously said Tesla could add traditional controls if necessary.Production should increase later this year, according to the report. Tesla will need enough vehicles to support a broader robotaxi network after completing its early Austin deployment.Meanwhile, the company continues to collect operational data from its existing robotaxi service. Those vehicles operate within selected areas where Tesla has established specific service boundaries.Waymo Expands its Driverless Ride NetworkTesla enters the purpose-built robotaxi market while Waymo expands its autonomous ride-hailing business. The Alphabet-owned company currently operates driverless services across several American markets.California recently approved Waymo’s expansion across the San Francisco Bay Area and Los Angeles. The company has also entered Sacramento and San Diego with its autonomous vehicles.Waymo now provides more than 500,000 fully autonomous electric rides each week. Its vehicles have also completed over 220 million rider-only autonomous miles.By comparison, Tesla has reported about 380,000 driverless miles across its robotaxi operations. The two companies use different vehicles, systems, and deployment strategies.Tesla currently uses modified passenger vehicles for its Austin robotaxi service. The Cybercab launch would add a vehicle built entirely around autonomous operation.Additionally, Tesla plans to expand its robotaxi services beyond Austin. Dallas and Houston appear among the next planned markets, although the company has not provided firm launch dates.Employee rides will mark the first stage of the Austin Cybercab launch. Tesla then plans to place the vehicles into its active robotaxi service within days.ALSO READ: Tesla, SpaceX Could Become One: What Elon Musk Stands to Gain from the DealJoin our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Reliance Jio is bringing back Jio Prime with a new benefit aimed at customers worried about rising mobile recharge costs. The membership, priced at a one-time Rs. 300, will allow eligible prepaid and postpaid users to protect their existing recharge prices until September 5, 2027.Enrollment for the revamped program is scheduled to begin on August 20, 2026. Unlike the earlier version of Jio Prime, the latest offering focuses heavily on price protection, giving subscribers a way to guard against possible tariff increases.The move comes at a time when India's telecom sector is facing expectations of another round of price revisions. Airtel already changed several prepaid plans, adding to speculation that other operators could follow with higher tariffs. Jio, however, did not announce a tariff hike alongside the return of Prime.How Jio Prime Price Protection WorksThe Rs. 300 membership is not tied to just one recharge plan. Jio said that the members can switch between eligible plans while retaining price protection, though the specific plans covered will depend on the program's terms.For example, if an eligible recharge currently costs Rs. 300 and later rises to Rs. 350, a Prime member could continue to use the protected price during the applicable period. The potential savings would therefore depend on how much Jio's tariffs eventually increase.Rs. 600 Voucher BenefitThe membership also comes with vouchers worth Rs. 600, although this is not direct cashback deposited into a customer's account. The benefit is split into two Rs. 300 vouchers. One voucher can be used when taking a new JioHome or JioPC connection, while another is linked to referring to a new Jio SIM. Prime members will also receive priority customer support and early access to selected products and trials.Also Read: Airtel vs Jio 84-Day Voice Plans: Which Recharge Offers Better Value for WiFi-First Users?Is Jio Prime Worth Rs. 300?For customers who regularly recharge their Jio number and expect to remain with the operator for the next year, the program could prove useful if tariffs increase significantly. However, the price-lock benefit is essentially a bet on future recharge prices. If Jio keeps tariffs unchanged or increases them only marginally, the financial advantage of paying Rs. 300 upfront becomes smaller.The membership is also optional. Customers who do not subscribe can continue using Jio's existing services without joining Prime. Ultimately, Jio Prime's return gives subscribers a choice: pay Rs. 300 now for greater certainty, or continue with regular recharges and accept whatever prices Jio sets in the future.Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Choosing a crypto platform in India is no longer just about trading fees or the number of coins available. Investors also need to evaluate regulatory compliance, custody, withdrawal access and tax records before depositing funds.India’s Financial Intelligence Unit (FIU-IND) updated its AML/CFT guidelines for virtual digital asset service providers in January 2026, reinforcing compliance obligations for platforms operating in the country. Here are seven checks Indian crypto investors should make.1. Check FIU-IND RegistrationConfirm whether the platform is registered with FIU-IND as a reporting entity. FIU registration means the provider falls under India’s anti-money-laundering framework and must follow applicable KYC, monitoring and reporting requirements. However, FIU registration should not be treated as a government guarantee against losses or hacking.2. Understand Who Controls the CryptoCheck whether assets are held in omnibus wallets, segregated custody or through an external custodian. Platforms should clearly explain their cold-storage practices and how withdrawals are authorized.Proof-of-reserves disclosures can provide additional information, but they do not, by themselves, prove that a company has no liabilities.3. Test Withdrawals Before Depositing MoreA platform is significantly less useful if users cannot move their crypto. Make a small deposit and test a withdrawal first. Check withdrawal fees, minimum amounts, supported networks and whether the exchange routinely places lengthy holds on transfers.Difficulty withdrawing funds is also a common financial-fraud warning sign highlighted in SEBI’s broader investor-safety guidance. 4. Look for Strong Account SecurityAt minimum, investors should look for two-factor authentication, withdrawal-address controls, login alerts and device management.Avoid relying solely on SMS authentication where stronger authenticator or hardware-key options are available.5. Verify the App and WebsiteFake investment applications remain a serious threat. SEBI warned in 2026 that cloned apps can closely imitate legitimate financial platforms. Install apps only from links verified on the company’s official website and independently check the domain before entering passwords or seed phrases.6. Check Whether Tax Records are DownloadableIndian investors need accurate transaction histories. The Income Tax Department states that VDA gains are taxed at 30%, plus applicable surcharge and 4% cess, and VDA income must be disclosed transaction-wise in Schedule VDA. A platform should therefore provide downloadable trade, deposit, withdrawal and TDS records.7. Be Suspicious of Guaranteed ReturnsGuaranteed profits, fixed high yields and pressure to deposit immediately are major warning signs. SEBI specifically advises investors to be cautious of guaranteed or unusually high returns and opaque investment schemes. Also Read: Best Crypto Hardware Wallets in 2026Final ThoughtsA polished app does not establish that a crypto platform is financially or operationally safe. Indian investors should evaluate compliance, custody, withdrawals, security and tax reporting together.The safest approach is also to avoid keeping more assets on an exchange than necessary when long-term holdings can be securely self-custodied.FAQs:1. Why should Indian investors check FIU-IND registration?FIU-IND registration shows that a crypto service provider falls under India’s AML and reporting framework. However, registration does not guarantee that a platform is financially safe or protected from hacks.2. Does proof of reserves mean a crypto exchange is completely safe?No. Proof of reserves can show that certain assets are held, but it may not provide a complete picture of liabilities, debts or operational risks. Investors should assess custody, withdrawals and transparency as well.3. Why should users test withdrawals before depositing large amounts?A small test withdrawal can reveal network support, fees, delays and account restrictions before more money is committed. It also confirms that users can actually move assets away from the platform when needed.4. What tax records should an Indian crypto platform provideInvestors should look for downloadable trade histories, deposits, withdrawals and applicable TDS records. Complete records are important as VDA income must be reported accurately for Indian tax purposes.5. What are common warning signs of an unsafe crypto platformGuaranteed returns, unusually high fixed yields, pressure to deposit quickly and difficulty withdrawing funds are major red flags. Fake apps, cloned websites and unclear custody arrangements should also be treated cautiously.

India’s mutual fund industry reached a record Rs 85.76 lakh crore in assets during July 2026. A year earlier, assets stood at Rs 75.36 lakh crore. Franklin Templeton’s industry dashboard reported the figures. Equity funds, systematic investment plans, and passive products supported the expansion. Investor participation also increased outside India’s largest financial centres.Meanwhile, JioBlackRock Asset Management plans to introduce ‘regular plans’ for eligible mutual fund schemes. Registered distributors will offer these plans to investors. The company currently sells only direct plans through its digital channels. Distributor commissions will add costs for investors choosing the new option.Equity Funds Lead Industry GrowthEquity mutual fund assets reached about Rs 52.3 lakh crore in July. They stood at Rs 45.1 lakh crore one year earlier. Equity funds now represent 61% of India’s mutual fund industry. Their share stood at 59.9% during July 2025.Equity schemes received Rs 24,697 crore in net inflows during the month. Small-cap funds attracted Rs 7,768 crore, while mid-cap funds received Rs 6,192 crore. Flexi-cap schemes recorded Rs 4,709 crore. Large and mid-cap funds added Rs 3,425 crore. Multi-cap funds drew another Rs 3,227 crore.Individual investors hold about 87% of their mutual fund assets in equity-oriented schemes. Institutions place more money in liquid, debt, and money-market funds. Individual investor assets grew at a 23% annual rate over five years. Overall industry assets grew at a 19% rate.SIP and Passive Fund Assets RiseMonthly SIP investments increased 12% from a year earlier to Rs 31,961 crore. July 2025 flows totalled Rs 28,464 crore. The number of SIP accounts rose 12.5% to 10.63 crore. New registrations reached 61.44 lakh during July.The average monthly SIP contribution edged lower to Rs 3,007 from Rs 3,012. The dashboard linked higher account closures partly to inactive-account reconciliation. Registrars, transfer agents, and exchanges carried out that process.Passive fund assets rose 24% to Rs 15.15 lakh crore. They totalled Rs 12.18 lakh crore one year earlier. Passive products now represent 18% of mutual fund assets. Their share stood at 15% in July 2022.Domestic equity products accounted for 68.4% of passive assets. Commodity funds held 16.5%, while debt products represented 12.4%. International funds accounted for 2.5%. Passive schemes formed 37% of all mutual fund schemes, compared with 34% last year.JioBlackRock Expands Distributor AccessJioBlackRock will use registered distributors to sell its regular plans. Direct plans exclude distributor commissions and usually carry lower costs. Regular plans include those commissions within their expense structures. Investors can choose between the available distribution routes.The company expects distributors to support complex and higher-value products. Managing Director and Chief Executive Sid Swaminathan previously identified special investment funds as products requiring adviser support. The strategy expands JioBlackRock’s reach beyond its digital-first model.Jio Financial Services and BlackRock formed the asset management venture. JioBlackRock has gathered about Rs 18,000 crore in assets within a year. Its portfolio includes cash, debt-index, and active equity funds.Earlier in August, the partners launched the JioBlackRock Nifty 50 ETF. The product marked their entry into India’s exchange-traded fund market. Meanwhile, mutual fund assets from B30 cities reached 19% of total industry assets in July. That share stood at 16% in December 2020.ALSO READ: India Needs 14.2% Rupee Growth to Reach $20 Trillion by 2036Join our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp

Apple will redesign its tracking pop-up after Germany’s Federal Cartel Office (FCO) ruled against its current design. The decision came on August 17, 2026, after a multi-year review of Apple’s App Tracking Transparency system. The regulator found that Apple’s own apps received more favorable consent treatment than third-party apps. Apple agreed to change the prompts within four months after receiving the formal decision. The German antitrust ruling targets how Apple presents tracking requests on iPhones and iPads. Third-party apps previously faced wording and visuals that could discourage users from granting permission. Apple’s own apps used more favorable consent messages, according to the FCO. The redesigned Apple privacy prompts must use neutral language and visuals across different app providers. The FCO said the issue involved unequal treatment rather than the existence of tracking controls. Its earlier assessment found Apple's approach could steer users differently depending on the app requesting consent. The authority described the concern as possible unequal treatment and self-preferencing under competition rules.Apple will also give developers more freedom around consent requests. Third-party publishers can combine Apple’s mandatory notice with separate privacy consent messages. This change could help advertising-supported apps explain data use more clearly before seeking permission. The new EU tracking rules will apply across almost all European Union countries, not only Germany. Apple must complete the changes within four months, while an independent trustee will monitor compliance for seven years. Apple has faced wider European pressure over its tracking framework. France previously fined Apple €150 million, while Italy imposed a €98.6 million penalty. The German decision adds another major challenge to Apple’s privacy strategy and advertising rules. Also Read: Apple Tests AirPods-Like Features for Third-Party Devices Amid EU PressureJoin our WhatsApp Channel to get the latest news, exclusives and videos on WhatsApp