Category: Business

Researchers are tracking a persistent web skimming campaign that targets major payment networks, using malicious JavaScript to steal credit card information from unsuspecting online shoppers.

As online shopping becomes increasingly familiar and convenient, a hidden threat lurks beneath the surface. Researchers are monitoring a long-running web skimming campaign that specifically targets businesses connected to major payment networks. This technique enables criminals to secretly insert malicious code into checkout pages, allowing them to capture payment details as customers enter them. Often, these attacks operate unnoticed within the browser, leaving victims unaware until unauthorized charges appear on their statements.

The term “Magecart” refers to various groups that specialize in web skimming attacks. These attacks primarily focus on online stores where customers input payment information during the checkout process. Rather than directly hacking banks or card networks, attackers embed malicious code into a retailer’s checkout page. This code, typically written in JavaScript, is a standard programming language used to enhance website interactivity, such as managing forms and processing payments.

In Magecart attacks, criminals exploit this same JavaScript to covertly capture card numbers, expiration dates, security codes, and billing details as shoppers input their information. The checkout process continues to function normally, providing no immediate warning signs to users. Initially, Magecart referred specifically to attacks on Magento-based online stores, but the term has since expanded to encompass web skimming campaigns across various e-commerce platforms and payment systems.

Researchers indicate that this ongoing campaign targets merchants linked to several major payment networks. Large enterprises that depend on these payment providers face heightened risks due to their complex websites and reliance on third-party integrations. Attackers typically exploit overlooked vulnerabilities, such as outdated plugins, vulnerable third-party scripts, and unpatched content management systems. Once they gain access, they inject JavaScript directly into the checkout flow, allowing the skimmer to monitor form fields associated with card data and personal information. This data is then quietly transmitted to servers controlled by the attackers.

To evade detection, the malicious JavaScript is often heavily obfuscated. Some variants can even remove themselves if they detect an admin session, creating a false impression of a clean inspection. Researchers have also noted that the campaign utilizes bulletproof hosting services, which ignore abuse reports and takedown requests, providing attackers with a stable environment to operate. Because web skimmers function within the browser, they can circumvent many server-side fraud controls employed by merchants and payment providers.

Magecart campaigns simultaneously impact three groups: the online retailers, the customers, and the payment networks. This shared vulnerability complicates detection and response efforts.

While consumers cannot rectify compromised checkout pages, adopting a few smart habits can help mitigate exposure, limit the misuse of stolen data, and facilitate quicker detection of fraud. One effective strategy is to use virtual and single-use cards, which are digital card numbers linked to a real credit or debit account without revealing the actual number. These cards function like standard cards during checkout but provide an additional layer of security. Many people can access these services through their existing banking apps or mobile wallets, such as Apple Pay and Google Pay, which generate temporary card numbers for online transactions.

A single-use card typically works for one purchase or expires shortly after use, while a virtual card can remain active for a specific merchant and be paused or deleted later. If a web skimming attack captures one of these numbers, attackers are generally unable to reuse it elsewhere, significantly limiting financial damage and making it easier to halt fraud.

Transaction alerts can notify users the moment their card is used, even for minor purchases. If web skimming leads to fraudulent activity, these alerts can quickly reveal unauthorized charges, allowing cardholders to freeze their accounts before losses escalate. For instance, a small test charge of $2 could indicate fraud before larger transactions occur.

Using strong, unique passwords for banking and card portals can also reduce the risk of account takeovers. A password manager can assist in generating and securely storing these credentials. Additionally, individuals should check if their email addresses have been compromised in past data breaches. Many password managers include built-in breach scanners that alert users if their information appears in known leaks. If a match is found, it is crucial to change any reused passwords and secure those accounts with new, unique credentials.

Robust antivirus software can block connections to malicious domains used to collect skimmed data and alert users about unsafe websites. This protection is essential for safeguarding personal information and digital assets from potential threats, including phishing emails and ransomware scams.

Data removal services can also help minimize the amount of personal information exposed online, making it more challenging for criminals to match stolen card data with complete identity details. While no service can guarantee complete data removal from the internet, these services actively monitor and systematically erase personal information from numerous websites, providing peace of mind and reducing the risk of targeted attacks.

Regularly reviewing financial statements, even for small charges, is another prudent practice, as attackers often test stolen cards with low-value transactions. The Magecart web skimming campaign illustrates how attackers can exploit trusted checkout pages without disrupting the shopping experience. Although consumers cannot fix compromised sites, implementing simple safeguards can help reduce risk and facilitate early detection of fraud. Online payments rely on trust, but this campaign underscores the importance of pairing that trust with caution.

As awareness of web skimming grows, consumers may find themselves reconsidering the safety of online checkout processes. For further information and resources on protecting against these threats, visit CyberGuy.com.

The U.S. Trade and Development Agency has announced support for a submarine cable project linking India and Singapore, aimed at enhancing connectivity and security in Southeast Asia.

WASHINGTON, DC – On January 20, the U.S. Trade and Development Agency (USTDA) announced its backing for a proposed submarine cable system that will connect India with Singapore and key data hubs across Southeast Asia.

The planned cable route is set to link Chennai, India, with Singapore, while additional landing points are under consideration in Malaysia, Thailand, and Indonesia, according to USTDA.

As part of this initiative, USTDA has signed an agreement with SubConnex Malaysia Sdn. Bhd. to fund a feasibility study for the SCNX3 submarine cable system. This project is expected to serve approximately 1.85 billion people by enhancing digital infrastructure in the region.

The feasibility study aims to attract investment for the cable system and expand the capacity necessary for Artificial Intelligence and cloud-based services. USTDA emphasized that this effort will also help ensure the reliability and security of international networks while minimizing exposure to cyber threats and foreign interference.

The agreement was formalized during the Pacific Telecommunications Council 26 conference held in Honolulu, Hawaii.

SubConnex has appointed Florida-based APTelecom LLC to conduct the feasibility study. The study will encompass various aspects, including route design, engineering, financial modeling, commercialization planning, and regulatory analysis.

The SCNX3 submarine cable is designed to address the increasing connectivity challenges faced by India and Southeast Asia. USTDA noted that the rising demand for digital services, coupled with limited route diversity, has rendered existing networks susceptible to outages and security vulnerabilities.

By introducing new and resilient data pathways, the project is anticipated to enhance digital access and support the growth of Artificial Intelligence and cloud services. USTDA stated that the cable will provide a secure and reliable communications infrastructure for governments, businesses, and citizens throughout South and Southeast Asia.

Furthermore, USTDA highlighted that the feasibility study will promote the use of secure cable technology, safeguarding data flows from potential malicious foreign influences. This concern is increasingly relevant as undersea cables facilitate the majority of global internet and data traffic.

According to IANS, the initiative represents a significant step toward improving digital connectivity in the region.

OpenAI is set to introduce advertising in ChatGPT for U.S. users on its free and Go-tier plans, marking a significant shift in its revenue strategy.

OpenAI is preparing to test advertisements within ChatGPT, targeting users of its free version and the newly launched Go-tier plan in the United States. This initiative aims to alleviate the financial pressures associated with developing and maintaining advanced artificial intelligence systems.

The company announced on Friday that the ads will begin appearing in the coming weeks, clearly distinguished from the AI-generated responses that users receive. Users subscribed to OpenAI’s higher-tier plans—Plus, Pro, Business, and Enterprise—will not encounter these advertisements.

OpenAI emphasized that the introduction of ads will not affect the quality or integrity of ChatGPT’s responses. Furthermore, user conversations will remain confidential and will not be shared with advertisers.

This move represents a significant shift for OpenAI, which has primarily relied on subscription revenue up to this point. It also highlights the increasing financial challenges the company faces as it invests billions in data centers and prepares for a highly anticipated initial public offering.

Despite currently operating at a loss, OpenAI has projected that it will spend over $1 trillion on AI infrastructure by 2030. However, the company has yet to disclose a detailed plan for funding this extensive expansion.

Industry analysts suggest that advertising could become a vital new revenue stream for ChatGPT, which currently boasts approximately 800 million weekly active users. Nevertheless, they caution that this strategy carries inherent risks, including the potential to alienate users and diminish trust if the ads are perceived as intrusive or poorly integrated.

“If ads come off as clumsy or opportunistic, people won’t hesitate to jump ship,” warned Jeremy Goldman, an analyst at Emarketer. He noted that alternatives like Google’s Gemini or Anthropic’s Claude are readily available to users seeking ad-free experiences.

Goldman also indicated that OpenAI’s decision to incorporate ads could have broader implications for the industry, compelling competitors to clarify their own monetization strategies, particularly those that promote themselves as “ad-free by design.”

OpenAI has assured users that advertisements will not be displayed to individuals under the age of 18 and that sensitive topics, such as health and politics, will be excluded from advertising content.

According to the company, ads will be tested at the bottom of ChatGPT responses when relevant sponsored products or services align with the ongoing conversation. This approach aims to ensure that advertisements are contextually appropriate and minimally disruptive.

Advertisers are increasingly optimistic about AI’s potential to enhance results across search and social media platforms, believing that more sophisticated recommendation systems will lead to more effective and targeted advertising.

Additionally, OpenAI confirmed that its ChatGPT Go plan, initially launched in India, will soon be available in the U.S. at a monthly subscription price of $8.

This new advertising initiative marks a pivotal moment for OpenAI as it seeks to balance user experience with the need for sustainable revenue growth, navigating the challenges of an evolving digital landscape.

For more details, refer to American Bazaar.

Google has launched a new Market Access Program aimed at helping Indian AI startups scale globally, coinciding with the projected growth of India’s AI market to $126 billion by 2030.

With India’s artificial intelligence (AI) market projected to reach $126 billion by 2030, Google has introduced a new Market Access Program designed to assist Indian AI startups in scaling their operations and expanding into global markets.

Announced during the Google AI Startups Conclave in New Delhi, the program aims to support startups from their initial seed stage to full-scale operations. Preeti Lobana, Vice President and Country Manager for Google India, emphasized the importance of this initiative, stating, “If you solve for India, you build for the world. Our focus now is accelerating how quickly Indian startups can scale, reach global markets, and deliver outcomes.”

Lobana noted that India’s AI startup ecosystem is entering a transformative phase, moving from prototypes to market-ready products and transitioning from early traction to sustainable business models. Google’s comprehensive support for startups encompasses capability building, real-world deployment, and scaling, addressing challenges at every critical stage of development.

The Market Access Program is specifically tailored for AI-first startups that are prepared to scale responsibly. It focuses on three key outcomes: enhancing enterprise readiness through global selling expertise, providing access to Google’s extensive enterprise network, and facilitating global immersion in key international markets.

To bolster the capabilities of these startups, Google also announced the upcoming Global AI Hub in Visakhapatnam. This facility, which will be powered by green energy and feature 1-gigawatt computing resources, is designed to equip startups with the high-performance computing necessary to refine their AI models on a global scale.

In addition to the Market Access Program, Google unveiled new updates to its Gemma model family, specifically targeting areas of rapid adoption in India, such as population-scale healthcare AI and action-oriented, on-device agents. The latest iteration, MedGemma 1.5, enhances Google’s health-focused AI initiatives by enabling developers to create applications that support complex medical imaging workflows.

The release of MedGemma 1.5 follows a collaboration between Google and the All India Institute of Medical Sciences (AIIMS), which is utilizing the model to develop India’s Health Foundation Models. This partnership contributes to the country’s Digital Public Infrastructure and enhances health outcomes across the ecosystem.

To support the growing demand for agent-based systems, Google introduced FunctionGemma, a specialized version of the Gemma 3 model. FunctionGemma is designed for function calling, allowing startups to translate natural language commands into executable actions. This capability enables the development of on-device, low-latency applications with automated workflows that prioritize user privacy and can function effectively on low-end devices without a constant internet connection.

Together, these advancements expand the toolkit available to Indian founders, facilitating the transition from experimentation to deployment across healthcare, enterprise, and consumer applications at scale. Lobana highlighted that these models are supported by popular tools throughout the development workflow, including Hugging Face Transformers, Unsloth, Keras, and NVIDIA NeMo.

Alongside the Conclave, Inc42 released the “Bharat AI Startups Report 2026,” which was supported by Google. The report reveals a significant shift in the AI ecosystem, with 47% of enterprises already moving from pilot projects to full production. It also notes a decrease in innovation costs, as historically high computing expenses have hindered Indian startups. With public resources lowering entry barriers, funding is increasingly directed toward product innovation rather than infrastructure costs.

India’s unique challenges, including its 22 languages, inconsistent connectivity, and price sensitivity, have often been viewed as obstacles. However, the report reframes these challenges as assets, suggesting that if an AI solution can effectively serve rural users in India, it is robust enough for global markets. The concept of “Bharat-tested” technology is emerging as a new benchmark for resilience.

The competitive landscape is shifting towards trust-by-design, with startups that prioritize safety, privacy, and security from the outset gaining a significant advantage in securing long-term enterprise contracts.

Ultimately, the success of AI initiatives will be measured by their outcomes. Examples include Cloudphysician, which has reduced ICU mortality rates by 40%, and Rocket Learning, which personalizes education for millions of students. Lobana concluded, “By stitching together skilling, capital, infrastructure, and market access, we are clearing the path for founders. As we look to the AI Impact Summit in February, the signal is clear: The future of AI isn’t just being used in India; it is being built here.”

According to Inc42, the launch of the Market Access Program marks a pivotal moment for Indian AI startups, positioning them to thrive in a rapidly evolving global landscape.

CloudSEK, an Indian cybersecurity firm, has secured a $10 million investment from Connecticut Innovations, marking a significant milestone as the first Indian-origin cybersecurity company to receive funding from a U.S. state fund.

CloudSEK, a Bengaluru-based cybersecurity firm specializing in predictive cyber threat intelligence, has announced a strategic investment of $10 million from Connecticut Innovations (CI), the venture capital arm of the State of Connecticut. This funding is part of the company’s Series B2 round and positions CloudSEK as the first Indian-origin cybersecurity company to receive backing from a U.S. state-backed venture fund.

The investment follows CloudSEK’s previous fundraising efforts, which included $19 million raised in its Series B1 round. With the completion of this latest tranche, the company has successfully concluded its Series B funding round, which consists of both primary and secondary capital.

“Becoming the first Indian-origin cybersecurity company to receive backing from a U.S. state fund is a milestone for CloudSEK, as well as for the entire Indian cybersecurity ecosystem,” said Rahul Sasi, co-founder and CEO of CloudSEK. He emphasized the significance of this achievement for the company and the broader Indian cybersecurity landscape.

With Connecticut serving as its U.S. anchor, CloudSEK is committed to job creation, localized research investment, and enhancing cyber resilience in the Western world. Sasi expressed pride in advancing the company’s identity as a truly Indo-American cybersecurity firm.

The partnership with CI was established after CloudSEK distinguished itself as a leading startup at VentureClash, CI’s global investment pitch competition. “At our 2025 VentureClash India pitch event, CloudSEK distinguished itself as a truly innovative provider of cybersecurity and predictive threat capabilities used by hundreds of businesses around the world,” stated Alison Malloy, Managing Director of Investments at Connecticut Innovations.

CloudSEK plans to utilize this investment to accelerate its expansion in the U.S., with intentions to establish a regional hub for operations, talent, and partnerships in Connecticut. The company aims to onboard strategic local talent and forge collaborations with corporate partners, universities, and research institutions throughout the state.

The funding from CI will enable CloudSEK to recruit top-tier cybersecurity and AI talent from the region, establish partnerships with local academic and research institutions, build its U.S. headquarters in Connecticut, and drive region-specific cybersecurity research and innovation.

This landmark investment not only enhances CloudSEK’s global trajectory but also symbolizes the growing prominence of Indian cybersecurity innovation on the world stage. By solidifying its presence in Connecticut and continuing to expand globally, CloudSEK is well-positioned to bolster cyber resilience across continents and redefine cross-border technology collaboration.

Prior to this investment round, CloudSEK’s Series B1 was led by U.S.-based strategic investor Commvault, with participation from MassMutual Ventures, Inflexor Ventures, Prana Ventures, and Tenacity Ventures. Early investors, including the Meeran Family (Eastern Group), StartupXSeed, Neon Fund, and Exfinity Ventures, continue to support the company’s long-term growth.

In addition to this funding, CloudSEK recently announced a strategic partnership with Seed Group, a company of The Private Office of Sheikh Saeed bin Ahmed Al Maktoum, aimed at delivering predictive cyber intelligence and AI-attack detection capabilities to organizations across the UAE.

Founded in 2015 by Sasi, a cybersecurity researcher-turned-entrepreneur, CloudSEK has evolved from a research-first initiative into a leading cyber threat intelligence platform, serving over 300 enterprises across various sectors, including banking, financial services, insurance (BFSI), healthcare, technology, and government.

This investment marks a pivotal moment for CloudSEK and highlights the increasing collaboration between Indian tech firms and U.S. state-backed initiatives, paving the way for future innovations in the cybersecurity domain, according to The American Bazaar.

Walmart has appointed Shishir Mehrotra, CEO of Superhuman, to its Board of Directors as the retail giant prepares for an agentic AI future.

Walmart Inc. has announced the appointment of Shishir Mehrotra, an Indian American technology veteran and current CEO of Superhuman, to its Board of Directors. This move comes as the retail giant positions itself for an agentic AI future.

Mehrotra will contribute to both the Compensation and Management Development Committee and the Technology and eCommerce Committee, as stated by the Bentonville, Arkansas-based company.

Greg Penner, chairman of Walmart’s Board of Directors, expressed enthusiasm about Mehrotra’s addition, saying, “Our focus remains on serving customers through a people-led, tech-powered approach. Shishir’s background adds to our boardroom the insight of a proven builder, offering a distinguished track record scaling platforms relied upon by millions.”

Randall Stephenson, the lead independent director, echoed this sentiment, highlighting Mehrotra’s unique skill set. “Shishir brings a rare combination of technical depth and product leadership. He has helped create and scale platforms that unlock creativity and productivity for people and teams at global scale. We’re excited to welcome him to our Board,” he remarked.

In response to his appointment, Mehrotra stated, “I have long admired Walmart’s ability to innovate while staying true to its core values, and joining the Board as the company builds for an agentic AI future is a rare opportunity. This era is the most significant technological shift I’ve seen in my career, and I look forward to working with the team to shape the future for the millions of people Walmart serves.”

Mehrotra brings over 25 years of experience in the technology sector, with a proven track record of building category-defining platforms. Before his role at Superhuman, an email application designed for productivity enhancement, he co-founded Coda, a productivity and AI platform that successfully served millions of users and tens of thousands of teams.

Prior to founding Coda, Mehrotra held significant positions at YouTube, serving as both Chief Product Officer and Chief Technology Officer. During his tenure, he played a crucial role in transforming YouTube into the world’s largest video platform and one of Google’s most significant and rapidly growing businesses, catering to a new generation of creators.

Mehrotra holds a dual Bachelor of Science degree in mathematics and computer science from the Massachusetts Institute of Technology.

Walmart serves approximately 270 million customers and members each week across more than 10,750 stores and various eCommerce websites in 19 countries. The company reported a fiscal year 2025 revenue of $681 billion and employs around 2.1 million associates globally, according to the company’s release.

This strategic appointment reflects Walmart’s commitment to integrating advanced technology into its operations and enhancing customer service as it navigates the evolving landscape of retail.

According to The American Bazaar, Mehrotra’s expertise will be invaluable as Walmart continues to innovate and adapt in a rapidly changing market.

Recent survey findings reveal that Anthropic’s Claude is the most popular AI tool among U.S. professionals, surpassing competitors like ChatGPT and Google’s Gemini.

In the rapidly evolving landscape of artificial intelligence, a new survey sheds light on the preferences of U.S. professionals regarding workplace AI tools. While major tech companies are eager to promote their proprietary AI solutions, it appears that users are making their choices based on performance rather than corporate allegiance.

Conducted by Blind, an anonymous professional community platform, the survey indicates that Claude, developed by Anthropic, has emerged as the most widely used AI model in corporate environments. Surprisingly, Claude has outperformed more established competitors, including ChatGPT and Google’s Gemini. According to the survey, 31.7% of respondents reported using Claude as their primary AI tool at work, regardless of their employer’s preferences.

The survey collected responses from verified U.S.-based professionals during December, with a significant number identifying as software engineers. Participants sought AI assistance across various tasks, including debugging, system design, documentation, and content generation.

Despite Claude’s leading position, the survey reveals a more complex reality: professionals are not committing to a single AI model. Instead, many are curating personalized toolkits tailored to their specific needs. Vasudha Badri Paul, founder of Avatara AI, shared her experience, stating that her daily workflow involves multiple platforms. “I use Perplexity and Notebook LLM most frequently. For research and learning, I go to Claude and Gemini, while ChatGPT is my go-to for content,” she explained. Paul also incorporates Notion AI for organization, Sora for short video generation, Canva Magic Studio for graphics, and Gamma for slide decks.

This trend reflects a pragmatic approach among users, who are increasingly willing to switch between tools rather than remain loyal to a single ecosystem.

When it comes to coding, Claude’s advantages become particularly pronounced. The survey indicates that among developers, Claude excels in software development tasks. Many respondents highlighted its capabilities in writing and understanding complex code, an area where company-backed tools often face resistance. The survey found that 19.6% of professionals use ChatGPT, while 15% rely on Gemini. GitHub Copilot is close behind with 14.2%, and another 11.5% reported using Cursor.

The survey also explored preferences within companies that have their own AI products. At Meta, for instance, 50.7% of surveyed employees indicated that Claude was their preferred AI model, while only 8.2% reported using Meta AI. A similar trend was observed among Microsoft employees, where 34.8% favored Claude, narrowly ahead of Copilot at 32.2%, with ChatGPT trailing at 18.3%.

One key takeaway from the survey is that corporate backing does not necessarily guarantee employee loyalty. In an era where productivity is increasingly driven by AI tools, professionals are prioritizing effectiveness over brand allegiance.

Nitin Kumar, an app developer and solutions manager, noted the shift in his own AI stack over the past year. He stated, “Claude is definitely the most superior for software development.” Kumar recently canceled his ChatGPT Plus subscription, citing a lack of utility. However, he acknowledged that the AI landscape is still evolving, adding, “Gemini 3 Pro changed the game completely for non-coding uses.” He believes that coding capabilities are now nearly on par with Claude Opus 4.5.

Kumar’s insights reflect a broader trend of users experimenting with different tools and comparing version upgrades to find the best fit for their needs.

Interestingly, Google employees showed the strongest internal alignment, with 57.6% of those surveyed using Gemini as their primary AI model. However, this preference did not extend beyond Google’s offices, as only 11.6% of Amazon employees selected Gemini as their top choice. Amazon’s own AI tools, such as Amazon CodeWhisperer, received minimal traction, with just 0.7% of respondents indicating they used it.

Ultimately, the survey highlights a significant shift in how professionals engage with AI. Rather than adopting tools based on corporate mandates or branding, workers are choosing solutions that demonstrably enhance their speed, accuracy, and overall output. While Claude currently leads the pack, its dominance may not be permanent, but it has certainly established a measure of trust among users for now.

According to Blind, the findings underscore the importance of user experience in the competitive AI landscape.

Nvidia has entered a licensing agreement with Groq, acquiring its technology and key executives while allowing Groq to remain an independent entity.

Nvidia has announced a significant licensing agreement with the startup Groq, which includes the hiring of Groq’s CEO and other key executives. This development was detailed in a blog post by Groq, highlighting a trend where major tech companies engage with promising startups to leverage their technology and talent without outright acquisitions.

Groq is known for its specialization in “inference,” a process that involves artificial intelligence models responding to user queries after they have been trained. While Nvidia has established dominance in the AI training sector, it faces increasing competition from both established rivals and emerging startups like Groq and Cerebras Systems.

The agreement has been characterized by Groq as a “non-exclusive licensing agreement” for its inference technology. Groq emphasized that this partnership reflects a mutual commitment to enhancing access to high-performance, cost-effective inference solutions.

As part of this deal, Jonathan Ross, Groq’s Founder, and Sunny Madra, Groq’s President, along with other members of the Groq team, will transition to Nvidia to help advance and scale the licensed technology. Despite these changes, Groq will continue to operate independently under the leadership of Simon Edwards, who will assume the role of CEO.

A source close to Nvidia confirmed the agreement, although Groq has not disclosed any financial details related to the deal. Reports from CNBC suggested that Nvidia had considered acquiring Groq for $20 billion in cash, but neither company has commented on this speculation.

Bernstein analyst Stacy Rasgon noted in a recent client communication that antitrust concerns could pose a significant risk in this arrangement. However, by structuring the deal as a non-exclusive license, Nvidia may maintain the appearance of competition, even as Groq’s leadership and technical talent transition to Nvidia.

Groq has seen substantial growth, more than doubling its valuation to $6.9 billion from $2.8 billion since August of last year, following a $750 million funding round in September. The company distinguishes itself by not relying on external high-bandwidth memory chips, which has insulated it from the memory shortages currently affecting the global chip industry. Instead, Groq utilizes on-chip memory known as SRAM, which accelerates interactions with chatbots and other AI models, albeit at the cost of limiting the size of the models it can serve.

In the competitive landscape, Groq’s main rival is Cerebras Systems, which is reportedly planning to go public next year. Both companies have secured significant contracts in the Middle East, further solidifying their positions in the market.

Nvidia’s CEO, Jensen Huang, recently delivered his most important keynote address of the year, emphasizing the company’s strategy to maintain its leadership as the AI market transitions from training to inference.

This licensing agreement with Groq marks another strategic move for Nvidia as it seeks to bolster its capabilities in the rapidly evolving AI landscape, ensuring that it remains at the forefront of technological advancements.

For further details, refer to Reuters.

Starbucks has appointed Anand Varadarajan, a veteran of Amazon, as its new chief technology officer, effective January 19, 2026.

Starbucks announced on Friday that it has appointed Anand Varadarajan as its new chief technology officer (CTO). Varadarajan, who spent nearly 19 years at Amazon, most recently led technology and supply chain operations for the tech giant’s worldwide grocery stores business.

In a memo announcing the hiring, Starbucks CEO Brian Niccol praised Varadarajan’s expertise, stating, “He knows how to create systems that are reliable and secure, drive operational excellence, and scale solutions that keep customers at the center. Just as important, he cares deeply about supporting and developing the people behind the scenes that build and enable the technology we use.”

Varadarajan will officially begin his role on January 19, 2026, and will also serve as executive vice president. He takes over from Deb Hall Lefevre, the former CTO, who departed in September amid a $1 billion restructuring plan that included a second round of layoffs.

With a strong educational background, Varadarajan is an alumnus of the Indian Institute of Technology (IIT) and holds a master’s degree in civil engineering from Purdue University, as well as a master’s degree in computer science from the University of Washington.

During his tenure at Amazon, Varadarajan was recently elevated to oversee the worldwide grocery technology and supply chain organizations, which encompass both the company’s Fresh brand and Whole Foods. He reported directly to Jason Buechel, Amazon’s grocery chief and the CEO of Whole Foods.

At Amazon, Varadarajan was instrumental in implementing grocery technology innovations, including a pilot program that introduced mini robotic warehouses in Whole Foods supermarkets. This initiative enabled consumers to shop from both the in-store selection and products from Amazon’s broader inventory, which are not typically available at the organic grocer.

Starbucks is currently navigating a significant turnaround strategy under Niccol, who took over as CEO in September 2024. The company recently reported that its quarterly same-store sales returned to growth for the first time in nearly two years, according to CNBC. Additionally, holiday sales have shown strong performance this season, despite ongoing strikes by baristas.

A key component of Starbucks’ turnaround strategy is its hospitality platform, Green Apron Service, which represents the company’s largest investment in labor at $500 million. This program is designed to ensure proper staffing and enhance technology to maintain fast service times. It was developed in response to the growth in digital orders, which now account for more than 30% of sales, as well as feedback from baristas.

In a related development, Starbucks recently announced it would pay $35 million to more than 15,000 workers in New York City to settle claims that it denied them stable schedules and arbitrarily reduced their hours. This settlement comes amid a continuing strike by Starbucks’ union, which began last month in various locations across the U.S. This marks the third strike to impact the chain since the union was established four years ago.

As Starbucks moves forward with its strategic initiatives, Varadarajan’s extensive experience in technology and supply chain management is expected to play a crucial role in the company’s efforts to enhance operational efficiency and customer satisfaction.

According to CNBC, the company is focused on leveraging technology to improve service and address the challenges posed by labor disputes.

ChatGPT’s mobile app has surpassed $3 billion in global consumer spending, reflecting rapid adoption of AI technology and a strong subscription model since its launch in May 2023.

OpenAI’s ChatGPT mobile app has achieved a significant milestone, crossing $3 billion in global consumer spending. This figure highlights the rapid adoption of artificial intelligence and the effectiveness of subscription-driven growth.

As of this week, the ChatGPT mobile app has surpassed $3 billion in worldwide consumer spending on both iOS and Android platforms since its launch in May 2023. According to estimates from app intelligence provider Appfigures, a substantial portion of this growth—approximately $2.48 billion—occurred in 2025 alone. This marks a notable increase compared to the $487 million spent in 2024, showcasing the widespread acceptance of AI tools on mobile devices.

The ChatGPT app reached the $3 billion milestone in just 31 months, outpacing other major applications. For instance, TikTok took 58 months to reach a similar figure, while streaming services like Disney+ and HBO Max required 42 and 46 months, respectively. This rapid adoption underscores ChatGPT’s unique position in the mobile app market.

A significant portion of the spending is attributed to paid subscription tiers, such as ChatGPT Plus and ChatGPT Pro, which provide users with access to advanced features and the latest AI models. The app’s visibility in mobile app rankings has also increased, reflecting a growing consumer willingness to invest in AI-powered services. This achievement establishes ChatGPT as one of the most rapidly monetized AI applications in mobile history.

The $3 billion figure encompasses total spending on iOS and Android devices since the app’s initial launch. When it first debuted in May 2023, it was available exclusively on iOS.

ChatGPT is an AI language model developed by OpenAI that can comprehend and generate human-like text based on user prompts. It employs advanced machine learning techniques to perform a variety of tasks, including answering questions, writing content, translating languages, summarizing text, and assisting with coding.

The model has been integrated into various platforms, encompassing both web and mobile applications. It offers users free access alongside paid subscription options that provide enhanced capabilities. As a result, ChatGPT has rapidly emerged as one of the most widely utilized AI tools, reflecting the increasing demand for conversational AI across sectors such as education, business, entertainment, and everyday problem-solving.

The swift rise of the ChatGPT mobile app signifies a broader shift in consumer engagement with artificial intelligence, indicating a growing comfort with incorporating AI tools into daily life. Beyond impressive revenue figures, its success illustrates a larger trend toward mainstream adoption of AI-powered applications, where users increasingly recognize the value of conversational AI for productivity, creativity, and problem-solving.

This milestone also highlights the effectiveness of a subscription-based model for monetizing advanced AI services, demonstrating users’ willingness to invest in tools that enhance efficiency and provide innovative capabilities.

The app’s accelerated adoption compared to other major platforms reflects evolving expectations among mobile users and the distinct appeal of AI-driven experiences that deliver immediate, tangible benefits. Furthermore, this growth suggests a potential expansion of AI across various sectors, from education and entertainment to professional workflows, as accessibility and user familiarity continue to improve.

According to Appfigures, the success of ChatGPT’s mobile app is a testament to the increasing integration of AI into everyday life.

Supportiyo, co-founded by Ashar Ahmad, is transforming the home service industry by providing small businesses with an AI-driven digital workforce to enhance operational efficiency and reduce missed calls.

In an exclusive interview, Ashar Ahmad, co-founder and CEO of Supportiyo, discusses how the startup is revolutionizing operations for small businesses through applied artificial intelligence (AI).

Supportiyo, co-founded by Ahmad, is an applied AI startup focused on creating a digital workforce specifically for home service businesses. Unlike most AI tools that cater to large enterprises or technical users, Supportiyo aims to bridge the gap for small businesses that seek effective outcomes rather than complex tools.

The platform functions as a vertical AI phone agent for home service businesses, addressing one of the industry’s significant revenue leaks: missed calls. Supportiyo answers calls instantly, comprehends trade-specific language, manages customer objections, and books jobs directly into company calendars. This solution emerged from the collaboration between Ahmad, an AI engineer, and Ahmad M.S., a trades business owner who experienced firsthand the operational challenges faced by small businesses.

In the interview, Ahmad elaborated on Supportiyo’s mission and core purpose. “Supportiyo is an applied AI company building a digital workforce for home service businesses,” he explained. “Today, most advanced AI and automation tools are built for enterprises, engineers, or power users. Small business owners don’t want tools, workflows, or configuration platforms. They want work to get done.”

Ahmad emphasized that Supportiyo’s purpose is to transform existing AI capabilities into autonomous AI workers that take ownership of essential business functions. “These aren’t tools that merely assist people; they’re systems designed to actively perform work inside a business,” he noted. By identifying core workflows in home service businesses, Supportiyo creates AI workers capable of managing responsibilities from start to finish, delivering real return on investment without requiring business owners to learn new software or alter their operations.

When asked about the inspiration behind Supportiyo, Ahmad shared that the company was born out of a specific problem: missed calls. “As a builder and AI engineer, I saw how much capability already existed and how poorly it translated into real outcomes for small businesses,” he said. “When Ahmad, who was running a home service business at the time, became our first customer, the problem became very concrete. His business was losing revenue simply because calls were missed while technicians were in the field.”

Ahmad pointed out that the home services sector is one of the most underserved markets when it comes to technology solutions. While industries such as hospitality, banking, and education have access to various tools, home services have lagged behind. “Supportiyo exists to close the gap between modern technology and practical execution,” he added.

Supportiyo’s unique approach to trades businesses sets it apart from generic call-handling solutions. “We combine deep technical capability with real domain expertise,” Ahmad explained. “Most platforms give businesses ingredients—tools, workflows, prompts, and integrations—that owners are expected to assemble themselves. We take a different approach.” Instead of providing a kitchen full of tools, Supportiyo offers prebuilt, industry-specific AI workers that understand trade language, objections, scheduling logic, and operational nuances.

Feedback from early adopters has been overwhelmingly positive, with users expressing relief and trust in the system. An HVAC business owner noted that handling calls while working in the field was a significant challenge. After implementing Supportiyo, every customer was attended to and scheduled promptly, allowing the owner to step in only when necessary. A local food business shared that language barriers had previously hindered customer interactions, but Supportiyo learned their full menu and preferences, enabling smooth conversations and allowing the team to focus on their core work.

Ahmad highlighted that Supportiyo now manages close to 80% of inbound calls for some service business owners, providing them with more time to concentrate on growth. “Owners often describe Supportiyo not as software, but as an extra worker they can rely on,” he said.

When discussing how the AI handles objections and nuanced customer queries, Ahmad explained that the AI operates with full business context rather than relying on scripts or hardcoded prompts. “Each AI worker understands the specific business it represents, including services, pricing logic, availability, and policies,” he stated. This capability allows the AI to respond based on real business rules and past outcomes, ensuring accountability and effective resolution of customer inquiries.

Building Supportiyo has not been without its challenges. Ahmad noted that educating potential customers about AI’s capabilities is crucial before selling the product. “We first have to explain what AI can realistically do, what it replaces, and what outcomes owners should expect,” he said. Trust has also been a significant hurdle, as the AI category has been marred by flashy products that fail in real operations. Supportiyo addresses this by focusing on reliability, narrow responsibilities, and maintaining tight feedback loops with customers.

Ahmad described a typical customer journey, which has evolved from a hands-on onboarding process to a more streamlined experience. “Today, onboarding is fast and simple. A customer creates an account, selects their industry, connects their website, and activates an AI worker. Within minutes, calls are being handled,” he explained. For those seeking guidance, assisted onboarding allows customers to go live in under ten minutes. “The core principle is that the AI adapts to the business. The business does not adapt to the AI,” he added.

Looking ahead, Ahmad envisions Supportiyo becoming the default AI workforce for home service businesses within the next five years. “Platforms like Jobber and ServiceTitan helped move the industry from paper to software. Supportiyo moves it from software to autonomous AI workers,” he said. The goal is not to replace people but to alleviate operational burdens, allowing humans to focus on judgment, relationships, and growth. “Home services are just the beginning. The mission stays the same as we expand: applied AI that takes responsibility for real work and delivers measurable impact,” he concluded.

According to The American Bazaar, Supportiyo is poised to make a significant impact on the home service industry by providing small businesses with the tools they need to thrive in an increasingly competitive landscape.