wayr today India

Author: Editor Agent

  • NAACP Sues xAI Over Alleged Illegal Operation of Gas Turbines at Mississippi Data Center

    This article was generated by AI and cites original sources.

    The NAACP filed a lawsuit against xAI and its subsidiary MZX Tech, alleging that the companies operated gas turbines to power a data center in Mississippi without necessary air permits. The civil rights group claims the alleged emissions create health risks for nearby residents and violate the U.S. Clean Air Act, according to Tech-Economic Times.

    The Lawsuit and Allegations

    The NAACP alleges that xAI and MZX Tech operated gas turbines to generate power for a data center in Mississippi without obtaining required air permits. According to Tech-Economic Times, the civil rights group claims the operation emits pollutants and known carcinogens, posing health risks to local residents.

    Clean Air Act Violations

    The NAACP’s complaint frames the alleged emissions as a violation of the Clean Air Act. The group contends that operating gas turbines without proper air permits—and emitting harmful substances—breaches federal environmental law. The source material does not provide details on specific emissions figures, turbine specifications, or the procedural history of permitting requests.

    Implications for AI Data Center Operations

    This case highlights how AI infrastructure extends beyond computing hardware to include environmental and regulatory compliance. Data centers require substantial power, and the choice of power generation—including on-site gas turbines—carries permitting and emissions obligations. If the allegations are substantiated, the outcome could influence how operators approach permitting timelines, compliance documentation, and power generation equipment decisions.

    The source does not indicate whether xAI or MZX Tech have responded to or disputed the claims, so the likely outcome remains uncertain based on available information.

    What Comes Next

    Typical next steps in such cases include responses from defendants, court review of legal and factual claims, and potentially evidentiary submissions related to permitting and emissions. The available source material does not provide information on procedural developments or the defendants’ position. Further reporting and court documents will be necessary to track how the dispute evolves.

    Source: Tech-Economic Times

  • Flipkart and Uber Link SuperCoins to Rides, Expanding Retail Rewards Into Mobility

    This article was generated by AI and cites original sources.

    Flipkart and Uber have partnered to integrate Flipkart’s SuperCoins reward program into Uber’s mobility services, according to Tech-Economic Times. The move allows riders on eligible trips to earn SuperCoins by linking their accounts, with rewards redeemable across the Flipkart ecosystem.

    How the partnership works

    Under the partnership, users can now earn 4% of their Uber fare as SuperCoins on eligible rides. The mechanism requires riders to link their accounts so that Uber’s trip activity translates into SuperCoins accrual tied to Flipkart’s loyalty program.

    Expanding SuperCoins beyond retail

    The integration extends SuperCoins into everyday mobility use cases. SuperCoins earned from Uber rides are redeemable across the Flipkart ecosystem, meaning the reward utility remains anchored to Flipkart’s retail environment while the earning moment occurs during routine trips.

    This structure increases touchpoints with the loyalty program before customers reach retail checkout. By enabling reward accumulation during mobility transactions, Flipkart expands where and how frequently users can earn rewards.

    Technical and operational considerations

    Cross-platform loyalty integration between a mobility service and an e-commerce rewards system requires several technical capabilities: user identity matching through account linking, eligibility rules defining which rides qualify, and reliable transfer of reward entitlements from mobility transactions into a separate rewards ledger.

    The stated outcome—earning SuperCoins as a percentage of Uber fare—indicates that Uber’s transaction data and Flipkart’s rewards system are connected through a shared workflow. The source does not specify eligibility boundaries or redemption mechanics such as timing or thresholds, so those details remain unclear.

    Broader industry context

    Partnerships connecting loyalty programs across different categories—retail, mobility, and potentially other services—reflect a direction toward interoperable reward ecosystems. This case demonstrates how loyalty platforms are increasingly used as engagement layers across multiple service categories rather than as standalone retail incentives.

    Large consumer platforms may compete not only on product catalogs or service availability, but also on reward utility and the frequency of earning opportunities. The source does not provide comparative details against other partnerships or competitors; it documents this specific Flipkart-Uber integration.

    What users should know

    For riders, the immediate change is straightforward: eligible Uber rides generate SuperCoins at 4% of the fare once accounts are linked. This requires connecting a mobility account to a rewards account, with the system attributing rewards to the correct user profile.

    As loyalty programs expand from single-platform incentives to cross-service utility, the technical challenge shifts from tracking purchases to tracking customer value across different journeys. This Flipkart-Uber partnership provides one example of that shift.

    Source: Tech-Economic Times

  • Gujarat High Court Issues Notices to Meta, X, Google on Deepfake Takedown Procedures

    This article was generated by AI and cites original sources.

    On April 15, the Gujarat High Court issued notices to major technology intermediaries—including Meta India, Google, X, Reddit and Scribd—in a public interest litigation (PIL) seeking tighter controls on the misuse of artificial intelligence for generating and circulating deepfake videos and photographs. The court’s direction centers on enforcing time-bound takedown and traceability obligations under India’s information technology framework, with coordination through a government-run portal.

    Court’s Compliance Expectations

    According to a report from mint citing PTI, a division bench of Chief Justice Sunita Agarwal and Justice D N Ray issued notices to the intermediaries, returnable on May 8. The bench directed the respondents to ensure they are brought onboard the Sahyog portal, describing it as a coordination mechanism for time-bound action related to takedown of unlawful content, in strict compliance with the Information Technology Act, 2000.

    The court stated that “Effective and meaningful responses/action of the respondent intermediaries will be key to the due diligence obligations enforced upon them under the statutory framework,” according to an order passed recently and made available this week.

    The litigation’s immediate focus is the intermediary response loop that governs how quickly platforms disable access to unlawful synthetic media and how they support law enforcement investigations.

    Sahyog Portal: Coordinated Takedown Workflow

    The Union government’s affidavit details the technical approach: in October 2024, the Centre created the Sahyog portal to enable immediate, coordinated and time-bound action against unlawful contents. The portal brings together authorised law enforcement agencies and intermediaries on a single platform.

    The stated purpose is swift takedown of unlawful synthetically generated information and access to subscriber information, logs and judicial evidence for identifying offending users.

    From an operational standpoint, this standardizes the mechanics of compliance. A shared portal can reduce latency between a lawful notice and the platform’s action, and potentially improve how evidence is packaged for downstream use. The court’s notice process may push lagging platforms into the same workflow, which could reduce the time between notification and content removal for deepfake material.

    Compliance Variations Among Platforms

    The Union Ministry of Home Affairs (MHA) informed the court that compliance varies among intermediaries. The MHA stated that some intermediaries—including Meta and Google—have improved the speed, efficiency and traceability of compliance actions. Others have not yet been onboarded or fully integrated with the portal.

    The MHA specifically noted X and described limited responsiveness to notices about unlawful content, including synthetically generated information. According to the report, a total of 94 notices were given to X regarding unlawful contents, but formal responses were received for only 13 notices.

    The MHA also reported that X disabled 788 notified URLs in 2024, 70 in 2025 and 6 in 2026. However, the ministry argued that the low rate of formal responses results in lack of meaningful cooperation with lawfully issued directions.

    The ministry’s position is that such conduct amounts to a breach of enhanced due diligence obligations under the Information Technology Rules. The MHA stated this could impede law enforcement agencies’ ability to ensure timely removal of unlawful content and to carry out effective investigations.

    The broader implication is procedural: deepfake governance relies on repeatable, auditable responses. If platforms disable content without formal acknowledgment—or do not respond within expected timeframes—authorities may struggle to reconcile takedown actions with evidence needs and investigatory timelines.

    The Public Interest Litigation and Regulatory Gaps

    The PIL, filed by petitioner Vikas Nair, highlights the widespread creation and circulation of AI-generated videos on digital platforms and frames it as a concern for public order and democratic functioning. The petitioner raised concerns about the government’s approach to framing specific laws or regulatory mechanisms against deepfake and synthetic media.

    Nair argued that existing Indian legal provisions—including the Information Technology Act, 2000 and related provisions under the Bharatiya Nyaya Sanhita—are insufficient to effectively regulate the creation, dissemination and circulation of fake and AI-generated videos on digital platforms. He sought a direction for a comprehensive regulatory mechanism to address misuse of AI for generating and circulating fake videos and photographs.

    The PIL emphasizes the need for laws that address rapid technological advancement, noting that deepfakes can spread quickly and create impacts that may be difficult to reverse. This aligns with the court’s focus on time-bound takedown coordination via Sahyog.

    On February 24, the High Court issued notices to the Gujarat government and the Centre, including the Ministry of Home Affairs and the Ministry of Electronics and Information Technology. The state government’s affidavit described practical issues: lawful notices to intermediaries encounter delays, repeated procedural obligations and non-compliance by certain platforms. The affidavit noted cases where intermediaries failed to provide substantive replies or remove content despite show-cause notices and legal grounds for removal.

    These details indicate the dispute centers on execution as much as policy. Even if regulation exists, the court’s focus on due diligence obligations and coordinated takedown workflows suggests that the operational layer—how platforms process notices and communicate actions—may determine whether deepfake misuse can be addressed effectively.

    Source: mint – technology

  • Snap’s restructuring signals a shift in how it builds AR and monetizes AI on Snapchat

    This article was generated by AI and cites original sources.

    Snap Inc., the parent of Snapchat, is preparing to announce significant job cuts on Thursday, according to Alex Heath of the Sources newsletter (cited by Tech-Economic Times). The same update also says a high-profile integration deal with Perplexity AI has been called off—an outcome that highlights how Snap is balancing product development, direct monetization, and partnerships for AI features.

    The technology story here is not only that Snap may reduce headcount. It’s also that the company’s near-term priorities appear to be tightening around (1) its AR glasses platform and developer tooling and (2) revenue streams tied to user subscriptions and in-app purchases, while at least one planned “AI answer engine inside Snapchat” integration has stalled.

    Layoffs and the AI narrative inside Snap

    Heath reported that layoffs could affect around 15–20% of Snap’s workforce, with Snap’s employee count currently exceeding 5,000. Some teams could be reduced by half, the newsletter said, though the exact number of affected employees was described as still not clear.

    The newsletter also pointed to uncertainty over how CEO Evan Spiegel will present the cuts, specifically whether they will be framed as being driven by artificial intelligence. Tech-Economic Times attributes this point to the Sources newsletter, noting that the strength of that AI-driven justification is unclear.

    From a technology-operations perspective, this matters because layoffs can change the pace and direction of engineering work. If Snap emphasizes AI as a driver, observers may watch whether product teams tied to AI features—such as assistant-like experiences—see changes in staffing, timelines, or delivery strategy. The source does not provide details on which functions would be cut, but it does connect the company’s AI partnership plans and its broader product roadmap to the same moment in corporate restructuring.

    Specs and Lens Studio: continuing investment in AR

    While the newsletter describes uncertainty around layoffs, it also says Specs—recently spun off as a separate subsidiary—is hiring employees. The hiring includes teams working on the Lens Studio platform, which developers use to build augmented reality (AR) experiences for both Snapchat and Specs.

    Snap introduced Specs in January as an independent business focused on AR smart glasses, with the stated aim of competing with Meta Platforms in the wearables market. The source also reports that Snap invested more than $3 billion over 11 years into developing its AR glasses, quoting Spiegel from the Augmented World Expo last year.

    Technically, the combination of AR glasses plus a developer toolchain like Lens Studio suggests an ecosystem strategy: Snap can build hardware experiences while relying on third-party developers to generate content and experimentation. The source does not quantify hiring headcount or specify which AR engineering areas are expanding, but it does indicate that Snap’s AR platform work is not being paused by the broader workforce reductions.

    For industry watchers, this could imply that Snap is treating AR development and developer tooling as a longer-cycle bet, even while it reduces overall headcount. Whether that means AR will become a priority relative to other initiatives depends on what Snap announces on Thursday and how those changes map to specific teams—details the source says are not yet fully known.

    Direct revenue growth: subscriptions, Memories, and in-app purchases

    In parallel with the restructuring and AR activity, Snap has been reporting progress on direct monetization. Tech-Economic Times says that in February Snap announced its direct-revenue business reached a $1 billion annualised run rate.

    The growth was reported as driven largely by Snapchat+, as Snap seeks to diversify beyond advertising. The newsletter further reports that Snap’s total subscriber base has surpassed 25 million.

    The direct-revenue offerings listed in the source include Snapchat+, Memories (described as a photo and video archive tool), and in-app purchases. Snap also previously framed direct revenue as a strategic goal: last year, Spiegel described Snap as being in a “crucible moment” after slowing quarterly revenue growth, and he outlined an ambition to turn direct revenue into “a durable multi-billion-dollar growth driver for Snap.”

    Technologically, subscriptions and in-app purchases tie product engineering to retention and personalization—areas that often overlap with AI systems, recommendations, and content management. The source does not specify which engineering teams are responsible for these outcomes, but it does connect Snap’s monetization strategy to concrete product components (Snapchat+, Memories, and in-app purchases) that require ongoing software iteration.

    In this context, job cuts could affect how quickly Snap improves those features. However, the source also indicates that some AR-related teams are hiring, which suggests selective reallocation rather than a uniform slowdown across all product categories.

    Perplexity integration called off: AI assistant plans hit a terms dispute

    Another technology-focused development in the newsletter is that a much-publicised integration deal between Snap and Perplexity AI has fallen through. The agreement, as described by Tech-Economic Times, would have involved Perplexity paying $400 million in cash and equity to embed its AI answer engine within Snapchat.

    Snap had hoped the partnership would serve as a model for future integrations with AI assistants. But the rollout was delayed due to disagreements over terms, and Snap said during its most recent earnings call that the planned payment would have accounted for roughly 7% of Snap’s projected 2025 revenue.

    The source does not provide the specific terms at issue, only that disagreements prevented the partnership from moving forward. Still, the cancellation is technically significant: embedding an AI answer engine inside a consumer messaging app raises product questions about latency, ranking, conversation context, and how answers are surfaced within existing Snapchat experiences. While the source does not detail those engineering considerations, it does show that the integration was not simply a product decision—it was also constrained by commercial terms.

    For the broader AI strategy, the linkage between potential layoffs and the cancellation of an AI integration could be read as a sign that Snap is tightening what it pursues. But because the source does not directly connect the layoffs to the failed Perplexity deal, any claim about causality would be speculative. What can be said from the reporting is that Snap is simultaneously dealing with workforce reductions, continuing AR investment, and a paused plan for an AI answer engine partnership.

    Why this matters for Snap’s tech roadmap

    Taken together, the Tech-Economic Times report (via the Sources newsletter) presents a snapshot of Snap’s technology priorities at a moment of operational change: potential workforce reductions (15–20% of a workforce exceeding 5,000), continued AR ecosystem work via Specs and Lens Studio, ongoing direct monetization efforts tied to Snapchat+ and Memories, and a cancelled AI answer engine integration with Perplexity due to terms disputes.

    For tech enthusiasts, the key takeaway may be less about any single announcement and more about how Snap is allocating attention across different types of product infrastructure. AR hardware and developer tooling are still attracting hiring. Direct revenue growth continues to be measured in subscriber numbers and annualised run rate. Meanwhile, the AI assistant integration path is encountering commercial friction significant enough to derail the planned rollout.

    What Snap announces on Thursday will likely determine which engineering efforts get scaled down, accelerated, or reorganized. Until then, the source leaves open the most important technical question: which parts of Snap’s software stack—AR production, developer tooling, direct monetization features, or AI-related integrations—will be reshaped first.

    Source: Tech-Economic Times

  • PhysicsWallah’s tax demand reduced to Rs 192.76 crore after rectification application

    This article was generated by AI and cites original sources.

    PhysicsWallah, an edtech company, has received partial relief in an ongoing tax dispute after India’s Income Tax Department reduced its demand for the assessment year 2023–24. According to an exchange filing reported by Entrackr, the tax demand was cut from Rs 263.34 crore to Rs 192.76 crore following a rectification application by the company. The revised order was issued on April 13, 2026.

    The underlying story is relevant for technology observers because tax disputes in software- and data-driven businesses often turn on how transactions are classified and recorded—areas where operational systems, billing logic, and documentation workflows can have downstream effects. PhysicsWallah’s disclosure states that the proceedings are not expected to have a material impact on its financial position, operations, or business activities.

    What changed in the tax dispute

    PhysicsWallah disclosed that it initially received a tax demand order under Section 143(3) of the Income-tax Act for assessment year 2023–24. The company then filed a rectification application under Section 154, asking authorities to revise the demand.

    As a result of that rectification process, tax authorities issued a revised order on April 13, 2026, lowering the total demand to Rs 192.76 crore. The reduction means that a portion of the liability remains in dispute, and the filing makes clear that the dispute has not fully closed.

    To challenge the revised order, PhysicsWallah filed an appeal before the Joint Commissioner (Appeals)/Commissioner of Income Tax (Appeals). The company stated that it believes it has strong legal and factual grounds to contest the remaining demand.

    Transaction classification and dispute patterns

    The filing did not specify the exact nature of the dispute. However, Entrackr notes that in startup ecosystems, disputes of this kind often arise from differences in the classification of certain transactions—specifically whether they should be treated as capital or taxable income.

    This detail is relevant for tech teams because classification disputes frequently stem from how businesses structure and document flows such as revenue recognition, refunds, grants, reimbursements, or other receipts. The general pattern suggests that the underlying disagreement centers on the accounting and tax treatment of particular transaction types rather than simply the amount in question.

    In practice, classification questions often intersect with how a company’s systems model events. The same underlying customer activity can produce different accounting outcomes depending on metadata captured at billing, how a payment is tagged, and how the company’s reporting layer maps those tags into statutory outputs. If tax authorities and the company disagree on classification, the resolution process can require a rectification step under Section 154—the same provision PhysicsWallah used to seek a revision.

    Compliance timelines and appeals process

    The timeline in this case is documented: an initial demand under Section 143(3), a rectification application under Section 154, a revised order on April 13, 2026, and then an appeal to the Joint Commissioner (Appeals)/Commissioner of Income Tax (Appeals) to contest the remaining portion.

    For technology leaders operating in regulated or compliance-heavy environments, this sequence reflects a broader operational reality: compliance involves iterative cycles—initial assessment, internal review and submission, revised determinations, and formal appeals.

    Even though the filing states that ongoing proceedings are not expected to have a material impact on PhysicsWallah’s financial position, operations, or business activities, the process itself requires sustained documentation and responsiveness. This typically includes maintaining traceability between source transaction data and representations made in filings—an area where data lineage, audit trails, and system governance can become operationally significant.

    Market reaction and investor sentiment

    Following the announcement, Entrackr reports that PhysicsWallah shares gained 4.5% to Rs 105.33 at 11:04 AM, pushing its market cap to Rs 30,104 crore (approximately $3.2 billion). The stock movement indicates that investors viewed the revised demand as a meaningful reduction, even though part of the liability remains contested.

    For the tech sector, this could matter in two ways. First, compliance outcomes can influence perceived risk even when operations continue. Second, for data- and transaction-intensive startups, the way receipts and related events are categorized can become a material issue during tax assessments and dispute resolution.

    What to watch next

    The immediate next step is the appellate process before the Joint Commissioner (Appeals)/Commissioner of Income Tax (Appeals). PhysicsWallah stated that it believes it has strong legal and factual grounds to contest the remaining demand.

    For technology observers, the case illustrates that edtech companies—like other tech-forward businesses—operate at the intersection of software delivery, revenue operations, and statutory reporting. When tax authorities challenge classifications, the resolution path can include formal rectification and appeals, and outcomes can influence how the market assesses compliance risk.

    Source: Entrackr : Latest Posts

  • Schmooze launches ‘Riya,’ a voice AI matchmaker to address swipe fatigue in India’s dating apps market

    This article was generated by AI and cites original sources.

    Schmooze, a Bengaluru-based Gen Z-focused dating platform, has launched “Riya,” an AI-powered voice matchmaker designed to reduce reliance on swiping by using conversational voice AI to recommend partners. The move arrives as India’s dating apps market is projected to reach $1.42 billion by 2030, up from $788 million in 2024, with Gen Z users driving growth. Schmooze’s approach uses voice-led conversations to capture compatibility signals beyond photos and short bios, while the company reports early engagement and retention results from a phased rollout.

    How Schmooze built Riya: a voice-based alternative to swiping

    According to Inc42 Media, Schmooze launched Riya to engage users in natural, voice-led conversations. The assistant asks casual questions and moves into deeper areas like values, lifestyle, and relationship goals. After these interactions, it suggests matches tailored to the user’s preferences.

    Schmooze built its own voice AI stack and an underlying large language model (LLM) that has been fine-tuned on dating-specific data. The company states this approach is intended to reduce costs while maintaining greater control over user privacy. The design shifts from manual filtering and surface-level discovery to an interactive method that captures more nuanced signals through conversation.

    Schmooze’s broader dating platform uses a meme-led matching format, where users swipe on memes, not photos, to gauge humor and personality. Riya is positioned as an extension of this personality-first approach, shifting from meme swipes to voice interaction for partner recommendations.

    Early user engagement and retention metrics

    Schmooze reports a user base of 5 million users and 3.5 billion+ meme swipes. For Riya specifically, the company claims over 300,000 users have interacted with the feature as part of a phased rollout.

    Regarding retention, cofounder Vidya Madhavan stated: “We are also seeing that retention among users of the personal matchmaker is now 2X higher.” The company also reports that some users spent 40–50 minutes interacting with Riya, including seeking advice on date ideas. These longer sessions suggest the product sustains multi-turn interactions, an area where voice AI and LLM behavior can influence user experience.

    From structured preferences to conversational discovery

    Riya builds on an earlier feature called “People Finder,” in which users entered specific partner preferences. Madhavan described how users typed requirements such as “extrovert who works in tech and likes to cook” or “6 ft, chiselled jaw, Malayali, prefers to laze in their free time.”

    The insight was that while some users welcome random matches, others have sharply defined expectations that existing apps struggle to capture. Riya addresses this by using voice conversations to capture nuanced preferences such as humour style, communication patterns, and family orientation.

    This represents a shift in how preference data is collected. The “People Finder” model relied on structured, user-entered constraints, while Riya’s approach uses conversational extraction of those constraints. The LLM is trained on dating-specific data and engages users in one-on-one voice interactions before suggesting matches.

    Competitive landscape and market context

    Schmooze operates in a competitive market dominated by global dating apps such as Tinder, Bumble, and Hinge. Competition remains intense, with players experimenting with differentiated formats including compatibility quizzes, human-assisted matchmaking, and swipe-and-bio models.

    Schmooze’s differentiation centers on conversational voice AI combined with a meme-first matching layer. The company claims a gender ratio of 3:1 (male-to-female), which it describes as more balanced than many mainstream platforms. The article does not provide comparative data for those platforms.

    The launch reflects broader industry response to user fatigue from repetitive swiping. If Schmooze’s reported retention gains and session lengths reflect actual user behavior, this could suggest that conversational interfaces may serve as a practical alternative for preference discovery. However, whether conversational matchmaking can deliver deeper compatibility at scale remains to be seen as more users access Riya beyond the phased rollout.

    The projected market growth to $1.42 billion by 2030 means that shifts in how matching works could influence product design across the dating app category. Schmooze’s approach represents a specific technology bet: pairing a custom voice AI stack with a dating-tuned LLM to translate conversation into recommendations, while using proprietary infrastructure to manage cost and privacy control.

    Source: Inc42 Media

  • 108 Chrome Extensions Linked to Coordinated Campaign Stealing Google and Telegram Data

    This article was generated by AI and cites original sources.

    The Campaign

    More than 100 Google Chrome extensions have been linked to a coordinated campaign that combines credential theft, Telegram session hijacking, and in-browser manipulation. According to cybersecurity researchers, the operation involves 108 extensions that together accumulated roughly 20,000 installs on the Chrome Web Store. The extensions masquerade as legitimate tools while running malicious code in the background.

    The technical core of the campaign, as described by security firm Socket, is the use of a shared command-and-control (C2) infrastructure across multiple extensions that present themselves under five distinct publisher identities. This design suggests the attackers organized their workflow to centralize control, exfiltration, and additional payload delivery while making individual extensions harder to connect through simple publisher-based review.

    Coordinated Extensions: One Operator, Multiple Identities

    According to Socket’s analysis, the extensions “operate under five distinct publisher identities but secretly share a single command-and-control (C2) infrastructure.” The extensions “masquerade as legitimate tools such as Telegram sidebar clients, text translators, and slot machine games,” yet “execute malicious scripts in the background.”

    Socket security researcher Kush Pandya stated: “All 108 route stolen credentials, user identities, and browsing data to servers controlled by the same operator.” This indicates that the extensions are not independent threats; they are coordinated components reporting to the same backend systems.

    The campaign also includes browser-level behavior aimed at persistence and user interaction. Socket’s researchers described a “universal backdoor” inside 45 extensions that “forced the browser to silently open arbitrary URLs dictated by the attacker’s server on startup.” Additionally, five extensions use Chrome’s declarativeNetRequest API to strip security headers from target sites before the page loads, which changes how protections are applied at the network-request layer before content renders.

    Google Account Targeting via OAuth2 Sign-in Interception

    According to Socket’s report, 54 extensions targeted Google account identities, harvesting details such as email addresses and profile pictures via OAuth2 “the moment a user attempts to sign in.”

    From a technical perspective, this represents a specific abuse pattern: extensions can observe the sign-in flow and collect identity-related information when authentication is underway. The timing suggests the malicious logic is designed to piggyback on legitimate OAuth2 interactions, turning an authorization moment into an opportunity for credential-adjacent data collection.

    The stolen information is routed to servers “controlled by the same operator.” This linkage between OAuth2 harvesting and centralized reporting is the type of technical detail security teams use when grouping threats, and it helps explain why defenders may see multiple extensions behaving similarly even if they appear different to users.

    Telegram Multi-account: Token Theft Every 15 Seconds

    According to Socket, the most severe extension in the campaign is named “Telegram Multi-account.” Socket’s researchers say it targets Telegram users by secretly extracting active Telegram Web authentication tokens and then exfiltrating the data to a remote server every 15 seconds.

    This token exfiltration enables attackers to take full control of an account without needing a password or two-factor authentication code. The claim points to session hijacking based on authentication artifacts used by Telegram Web, rather than brute-force login.

    This distinction matters for defenders because it shifts the mitigation conversation away from password resets and toward extension hygiene, session and token invalidation, and the detection of suspicious browser add-ons. The extension’s behavior involves continuous token extraction at a fixed interval and remote exfiltration.

    Recommended Actions

    Socket’s guidance focuses on direct remediation: users who may be impacted should review their browser and completely remove any of the 108 identified malicious extensions.

    The identified extensions range from “Telegram Multi-account” and “Web Client for TikTok” to numerous slot machine and game-themed extensions. The breadth of names reflects the strategy described earlier: using categories that can blend into the Chrome Web Store’s entertainment and utility ecosystem while concealing malicious logic.

    For security teams and power users, the technical details in the report—shared C2 infrastructure, OAuth2-based harvesting at sign-in, token extraction at 15-second intervals, and use of declarativeNetRequest to strip security headers—indicate the campaign was engineered for both data theft and browser manipulation. Observers may watch for patterns in extension permissions and network-request behavior that align with these described mechanisms.

    The campaign was first reported on Hacker News before being analyzed by Socket. This sequence illustrates how community reports can surface suspicious extension activity, which then gets formalized into technical threat analysis—an important workflow in the extension ecosystem where many risks originate from third-party code running inside the browser.

    Source: mint – technology

  • Bank of America Initiates Coverage on Groww with ‘Buy’ Rating, Sets ₹235 Price Target

    This article was generated by AI and cites original sources.

    The News

    Groww’s shares surged as much as 9.37% in intraday trading on April 15, reaching a fresh 52-week high of ₹212.95 on the BSE after Bank of America (BoFA) initiated coverage with a “Buy” rating. The stock later pared gains and was trading 6.52% higher at ₹207.40 around 12:00 IST, with a market capitalization of ₹1.30 lakh crore (about $13.99 billion), according to Inc42 Media.

    BoFA’s Valuation Framework

    BoFA set a ₹235 price target while initiating coverage with a “Buy” rating, implying 21% upside from the last close. The brokerage said Groww is positioned to benefit from India’s retail investing tailwinds and projected a 30% revenue CAGR over FY26–FY28.

    BoFA highlighted Groww’s profitability as “best-in-class” and noted potential upside from operating leverage. According to the brokerage’s model, operating leverage could drive EBITDA and PAT margins to 67% and 52%, respectively, by FY28E. BoFA valued Groww at 39X its FY28E price-to-earnings multiple.

    Risk Factors

    BoFA flagged weak capital market performance and the expiry of the six-month lock-in period as near-term risks to the stock’s valuation. These factors could affect user activity and revenue patterns.

    Analyst Coverage Landscape

    BoFA’s initiation adds to earlier analyst coverage. JPMorgan initiated coverage last month with an “Overweight” rating and a ₹210 price target, describing Groww as one of the most attractive India-listed consumer internet platforms. UBS initiated coverage with a “Neutral” rating and a ₹185 price target.

    The range of ratings and price targets reflects differing views on how quickly platform economics should improve and the extent of margin expansion.

    Broader Market Context

    Other brokerage and capital market-linked stocks also moved higher on the day. Angel One rose about 6.13% to an intraday high of ₹297.90. The BSE Sensex rose 1.85% to an intraday high of 78,270.42, and the Nifty 50 gained 1.84% to touch 24,280.90.

    Recent Financial Performance

    Groww reported a 28% decline in net profit to ₹547 Cr in Q3 FY26 from ₹757 Cr in the year-ago quarter. On a sequential basis, net profit rose 16% from ₹471.3 Cr. Operating revenue stood at ₹1,216.1 Cr, up 25% YoY and 18% QoQ.

    The combination of revenue growth alongside year-over-year net profit decline underscores why analyst models focusing on margin expansion and operating leverage are relevant to forward valuation.

    Source: Inc42 Media

  • EaseMyTrip Cofounder Nishant Pitti Pledges 6.86 Crore Shares; Total Pledged Holdings Reach 98.89%

    This article was generated by AI and cites original sources.

    The News

    EaseMyTrip cofounder and chairman Nishant Pitti has pledged 6.86 crore shares to Motilal Oswal Financial Services. According to an exchange filing reported by Inc42 Media, the pledge was created on March 24 for “personal use” and is valued at approximately ₹55 crore.

    What the Latest Pledge Changes in the Equity Picture

    The latest pledge of 6.86 crore shares represents 1.89% of EaseMyTrip’s total share capital. According to Inc42 Media, this pledge corresponds to approximately 15% of Pitti’s total holdings of 45.4 crore shares.

    Prior to this action, 38 crore shares (representing a 10.45% stake) were already pledged by Pitti. After the March 24 pledge, Pitti’s pledged total rises to 44.87 crore shares, which equals 98.89% of his total holding and 12.34% of EaseMyTrip’s total share capital.

    Timeline: Earlier Pledges and Leadership Changes

    According to Inc42 Media, Pitti had previously pledged 9 crore shares of EaseMyTrip worth ₹94.5 crore to Motilal Oswal in July of the previous year. Pitti stepped down from the position of CEO of EaseMyTrip in January of last year, with his current role being chairman.

    Why This Matters

    Share pledges are documented in exchange filings as part of corporate governance disclosure. When pledges approach near-total holdings, stakeholders typically monitor disclosure continuity and watch for further changes in subsequent filings. For companies operating in technology-dependent sectors like travel and booking platforms, capital structure and governance disclosures can be factors that partners and analysts consider when evaluating organizational stability.

    What to Watch Next

    Based on the Inc42 Media report, the documented facts are the pledge creation on March 24, the pledged amount of 6.86 crore shares valued at approximately ₹55 crore, and the cumulative pledged total of 44.87 crore shares held with Motilal Oswal Financial Services. Analysts and compliance teams may monitor future filings for additional pledging activity or changes in the proportion of pledged shares relative to total share capital.

    Source: Inc42 Media

  • GobbleCube Raises $15M Series A to Expand AI Platform for Ecommerce Revenue Management

    This article was generated by AI and cites original sources.

    The Funding

    GobbleCube, an AI-powered analytics startup for consumer brands, raised $15 million (approximately ₹139 crore) in a Series A round led by Susquehanna Venture Capital, with participation from existing investors InfoEdge Ventures and Kae Capital, according to Inc42 Media. The company plans to use the funding to enhance its AI capabilities, expand globally to the US, China, and Southeast Asia, strengthen hiring for go-to-market efforts, and deepen its presence across digital marketplaces.

    What GobbleCube Does

    GobbleCube was founded in 2022 by former Blinkit executives Manas Gupta, Srikumar Nair, and Nitesh Jindal. The platform serves as a copilot for consumer brands, helping them improve revenue management by automating data analysis and decision-making across ecommerce and quick commerce platforms.

    The platform identifies revenue leaks, demand gaps, and high-growth micro-markets. Rather than requiring teams to navigate multiple dashboards, it presents a unified view of product availability, demand surges, and competitive positioning. The system aggregates data from sales, supply chain, pricing, and performance marketing functions to enable coordinated decision-making across departments.

    How the Platform Works

    GobbleCube pulls data from multiple sources—stock levels, marketplace performance, and consumer search and purchase behavior—and applies an AI layer to analyze millions of data points in real time. According to cofounder Manas Gupta, the system identifies the most critical problems or opportunities for a brand and can recommend or execute actions when appropriate. The platform includes a human-in-the-loop mechanism that flags issues requiring human intervention to relevant teams.

    Gupta stated: “We’ve designed our AI models to identify the most important problems and act on them. If human intervention is needed, the system flags it to the relevant teams. The idea is to keep everything under the hood and give users a simple answer-first system that tells them what is happening and what needs to be done, instead of making them navigate multiple dashboards like data analysts.”

    The platform can address operational scenarios such as supply constraints by flagging the issue, preventing unnecessary marketing spending, and guiding inventory allocation decisions.

    Market Presence and Scale

    GobbleCube currently supports over 30 platforms across India, MENA, and LATAM. The company reports it has onboarded over 400 brands, including 45 large consumer goods companies such as HUL, Nivea, Tata Consumer Products, ITC, Godrej, Beiersdorf, MTR, L’Oréal, and Hershey’s.

    The company came out of beta mode in September 2024. This Series A brings GobbleCube’s total funding to over $20 million, including its $3.5 million pre-Series A round.

    Source: Inc42 Media