you can search by category
or news title

Microsoft Introduces $99 Microsoft 365 E7 Bundle to Bring AI Features to Businesses

Posted on by Yashika Aneja

Redmond: Microsoft has introduced a new USD 99‑per‑user software bundle called Microsoft 365 E7 to help companies use AI more easily in their daily work. The bundle will be available starting May 1, 2026. It brings together several of Microsoft’s top products, including Microsoft 365 E5, the Copilot AI assistant, and Agent 365. 

Microsoft says the goal is to make AI simple to adopt by offering everything in one package instead of making businesses buy separate tools. The company’s Work IQ system powers the bundle, helping AI understand context from emails, documents, meetings, and chats so it can give more useful responses and help with real work. 

Judson Althoff, Microsoft’s commercial business CEO, called the launch part of a larger “Frontier Transformation”, that is to say, Microsoft wants AI to help companies grow, work smarter, and reach bigger goals, not just automate simple tasks. He said companies no longer want experimental AI; they want tools that create real results.  

A major addition in the bundle is Agent 365, priced separately at $15 per user. Microsoft describes it as a control center for AI agents. It lets IT and security teams monitor all the AI agents running in the company, making sure they are safe and behaving as expected. This matters now more than ever because many businesses are beginning to treat AI agents like a digital workforce that needs proper supervision. 

Elon Musk’s Grok Sparks Outrage After Posting Racist Replies 

Posted on by Yashika Aneja

London: Elon Musk’s AI chatbot Grok, built by xAI, is currently facing severe criticism online for its racist and offensive replies. A Sky News report shared by MSN found that Grok responded to user prompts with hateful and abusive language, raising serious concerns about how safely the AI behaves and how well the platform is being moderated.

The controversy started when users began sharing screenshots that showed Grok making derogatory comments about certain communities and groups. Reports said some of its replies included racial slurs, insults, and harmful stereotypes, which sparked public anger and demands for accountability. In response, X’s internal safety team launched an urgent investigation to figure out how Grok gave such replies and whether its weak moderation filters were behind the whole thing.

The incident adds to growing scrutiny over Grok, which has already drawn regulatory attention for producing inappropriate content in the past. Governments have been monitoring the bot closely, especially as AI‑generated material spreads rapidly online. In January, xAI was even forced to limit certain image‑editing features in regions where the content was considered illegal.

Critics say Grok’s design, allowing it to reply publicly on X rather than in private chat, makes harmful posts easier to spread. Policymakers in the UK have also condemned the replies, calling them deeply concerning. X has stated that it is investigating the matter and reviewing Grok’s behavior before allowing it to interact publicly again.

Anthropic’s Claude AI Uncovers 22 Security Flaws in Firefox

Posted on by Yashika Aneja

San Francisco: Anthropic’s AI model, Claude Opus 4.6, found 22 security flaws in the Firefox browser during a two‑week project with Mozilla. This showcases how AI tools are becoming increasingly useful in helping experts identify security problems more efficiently. 

According to Mozilla, the discovered issues include 14 high‑severity, 7 moderate, and 1 low‑severity vulnerability. Most of these flaws have already been patched in Firefox version 148, while the remaining ones are scheduled for upcoming releases. The vulnerabilities were uncovered between January and February 2026, when Claude scanned nearly 6,000 C++ files across Firefox’s codebase. 

Claude also made a big discovery early on. Within just 20 minutes, it found a serious bug in Firefox’s JavaScript engine called a use‑after‑free issue. Human experts then tested the bug in a safe, virtual setup to make sure the AI was correct and that it wasn’t making a mistake. Overall, Anthropic sent 112 bug reports to Mozilla. This helped Mozilla review and fix problems much faster, speeding up the entire security‑patching process for Firefox. 

Anthropic also checked whether Claude could actually use the bugs it found to create real attacks. They tried this several times and spent about $4,000 on API credits. In the end, Claude managed to turn only two of the vulnerabilities into working exploits. Researchers say this proves that finding bugs is much easier than turning them into real‑world attacks, even for advanced AI systems. 
 

Nitro Commerce Acquires Zodiac Labs to Boost AI Analytics

Posted on by Yashika Aneja

GURUGRAM: Enterprise tech startup Nitro Commerce has acquired Gurugram-based Zodiac Labs AI for an undisclosed amount. The deal significantly expands Nitro’s AI capabilities, adding a real-time agentic analytics copilot purpose-built for India’s fast-growing quick commerce ecosystem.

Founded by Nikhil Garg in 2023, Zodiac Labs develops AI-powered tools that help consumer brands track and grow their presence across quick commerce platforms. Its core product suite includes a real-time SKU monitoring dashboard, a smart purchase order engine, and an AI-driven budget allocation system.

The platform currently serves brands operating on Blinkit, Zepto, and Swiggy Instamart. It does not merely surface data, it actively nudges marketers, unblocks stalled purchase orders, and reallocates ad spend in real time based on live performance signals.

Nitro Commerce has historically focused on D2C and ecommerce marketing solutions, serving over 2,500 brands including Rare Rabbit, Dot & Key, Blue Tokai, VIP Industries, and Pepperfry. The Zodiac acquisition marks a deliberate push into quick commerce, extending Nitro’s footprint across ecommerce, retail, and rapid delivery channels in a single move. Garg joins Nitro as part of the transaction.

The deal follows Nitro Commerce’s $5 million funding round raised earlier this year to scale international operations. It arrives as investor appetite for quick commerce infrastructure intensifies.

OpenAI Delays ChatGPT Adult Mode to Focus on Core Features

Posted on by Shobhit Kalra

SAN FRANCISCO: OpenAI is pushing back the launch of “adult mode” for ChatGPT for the second time. The feature would give verified adult users access to erotica and other explicit content inside the chatbot.

CEO Sam Altman first announced the feature in October 2025. He promised it would roll out in December alongside broader age-gating systems. That deadline slipped after Altman reportedly sent an internal “code red” memo. He called on teams to refocus on the core ChatGPT experience instead.

The company now confirms a second delay. An OpenAI spokesperson tells Axios the company is stepping back to “focus on work that is a higher priority for more users right now.” That work includes improvements to intelligence, personality, and making ChatGPT more proactive in conversations.

“We still believe in the principle of treating adults like adults, but getting the experience right will take more time,” the spokesperson said.

No new launch timeline has been given. The delay signals that OpenAI is prioritizing broad user experience gains over niche feature rollouts. It also reflects pressure the company faces from rivals including Google Gemini and Anthropic’s Claude.

Adult mode was positioned as part of OpenAI’s push to reduce paternalistic content restrictions. The company has faced criticism from users who say ChatGPT is overly cautious. It remains unclear when the feature will return to the roadmap.

Pentagon Designates Anthropic a Risk, Prompting Legal Response

Posted on by Yashika Aneja

Washington, D.C.: The Pentagon has officially labeled Anthropic as a supply‑chain risk in what is being described as the first time a U.S. AI company has received such a tag. 

According to the report, the Department of Defense told Anthropic that the decision takes effect immediately. This means the government no longer sees the company as secure enough to work with on defense projects. 

This decision will significantly limit Anthropic’s ability to work with U.S. defense agencies, as the government now questions how safely the company handles sensitive technology. Officials say the label was influenced by Anthropic’s refusal to give the Pentagon unrestricted access to its AI systems. The company had pushed back earlier, warning that such access could lead to mass surveillance or even the development of autonomous weapons, something it strongly opposes. 

Following the designation, Anthropic’s CEO Dario Amodei said the company believes the decision is not legally justified. He stated that Anthropic now feels it has “no choice” but to challenge the Pentagon in court. Amodei also noted that while the designation has a “narrow scope”, it still places them in an unprecedented situation for a U.S. AI company.  

This conflict shows how tensions are rising between AI companies and the government, as the U.S. tries to balance new technology with national security. The result of this case could shape how future AI firms are regulated and how much access government agencies can demand from private tech companies. 

Meta Sued After Contractors Viewed Private Footage from Smart Glasses 

Posted on by Yashika Aneja

Meta is currently facing a lawsuit in the U.S. after reports showed that overseas workers were reviewing private footage from its AI‑powered smart glasses. A Swedish investigation found that contractors in Kenya viewed recordings containing nudity and other personal moments, raising concerns about how Meta manages sensitive user data. 

The lawsuit, filed by two U.S. consumers, argues that Meta misled buyers by promoting the glasses as “designed for privacy” and “controlled by you”. Users believed their recordings were secure, but the investigation revealed that footage from more than 7 million glasses sold in 2025 entered a review system with no option to opt out. 

Meta says content remains on the device unless shared with Meta AI. However, sources reported that the company’s face‑blurring tools often failed, allowing contractors to clearly see private moments. 

As a result, the U.K.’s Information Commissioner’s Office has opened an inquiry into how Meta takes care of footage from the glasses and whether or not they inform users about the possibility of humans reviewing their recordings in advance. 

The case sheds light on the broader issue of privacy in the world of AI, where people often fail to understand how their data is captured or used. The lawsuit, as such, increases pressure on Meta to strengthen its privacy practices. 

OpenAI Launches GPT‑5.4, an AI That Can Operate Computers 

Posted on by Yashika Aneja

San Francisco: OpenAI has introduced GPT‑5.4, a powerful new AI model that, unlike earlier models, can use a computer on its own, opening apps, clicking buttons, typing, and even understanding what is happening on the screen. 

The new model comes in two versions, namely GPT‑5.4 Thinking and GPT‑5.4 Pro. Both versions are programmed to make complex office work easy by handling tasks across spreadsheets, documents, presentations, and software tools. The model can read screenshots, send keyboard and mouse commands, and follow multi‑step instructions much like a human assistant. 

GPT‑5.4 also supports a massive 1 million‑token context window, that is to say, it can process very large files, entire codebases or long documents, without losing track. It combines major improvements in reasoning, coding, and long‑running task execution, making it much more reliable for professional use. According to OpenAI, it is 33% less likely to produce incorrect information compared to earlier versions. 

The update includes 6 major boosts, such as better coding abilities, improved image understanding, and stronger performance on tasks that involve multiple steps. It also performs well in areas like legal analysis, financial modelling, and creating slide decks, helping users complete detailed work in a quick and efficient manner. 

With GPT‑5.4, OpenAI moves beyond AI that only answers questions. For this new model can now take action and boost your workplace productivity like never before. 

Rozana Secures Rs 290 Crore to Expand Rural Retail Platform

Posted on by Shobhit Kalra

Bengaluru: Rural commerce startup Rozana has raised Rs 290 crore in a Series B funding round led by Bertelsmann India Investments. The round also saw participation from Fireside Ventures, Spark Growth Ventures, Bikaji Family Office, FE Securities, and several other family offices.

The company plans to use the funds to expand its presence in more regions, increase product offerings, build private-label brands, and strengthen its technology and supply chain systems.

“For the last two and a half years we have been in the same villages just trying to test the depth of the market,” said co-founder Ankur Dahiya. “Now we would like to expand our geography to the Gangetic belt beyond where we are operating today.”

Rozana currently serves over one million households across more than 21,000 villages in Uttar Pradesh and Haryana. Its operations are supported by a network of over 35,000 local delivery entrepreneurs.

The startup focuses on gram panchayat-level markets, where organised retail and e-commerce penetration is still limited. “We started from gram panchayats and built the model from there,” Dahiya said.

Customers can order through Rozana’s app or website, with deliveries handled by local partners or through 86 distribution centres that also function as retail stores.

“Today, across these 21,000 villages, we have around 35,000 active entrepreneurs who deliver products for us,” Dahiya said.

Rozana combines several village orders into a single delivery to improve logistics efficiency. “Our average order value is about Rs 584 right now but what matters more for us is the average order delivery value,” Dahiya said. “Normally this remains between Rs 2,500 and higher because we combine five or six orders and deliver them together,” he added.

The company also plans to expand its product catalogue beyond the current 2,300 SKUs. “In our opinion, 2,300 SKUs is still a very small number. This will have to grow to probably 15,000 SKUs if we want to serve all the categories that the consumer buys,” Dahiya said.

Tata Elxsi Introduces DevStudio.ai for Faster Car Software Development

Posted on by Yashika Aneja

Bengaluru: Tata Elxsi recently unveiled DevStudio.ai, a new AI-powered tool programmed to help car companies build software in a quick and efficient manner. The platform makes use of multiple AI agents to attain the same and aid engineers, making the development process smoother. 

DevStudio.ai follows ASPICE standards, which are important markers of software quality and safety in the automotive industry. The tool also offers flexible deployment, that is to say, it can be used either in the cloud or on secure, on‑premise systems, depending on the company’s security needs. 

Unlike other general AI tools, DevStudio.ai is designed specifically for the needs of automotive software teams. It supports the development process in its entirety, right from early planning and design to testing and final validation. The tool also keeps everything connected with full traceability so teams can easily follow every step of their work. In addition to that, it also easily fits into the toolchains that car manufacturers and suppliers already use. 

Early use of DevStudio.ai by companies in North America, Japan, and India has already shown clear boosts in productivity and noticeably faster time‑to‑market, especially across areas like body electronics, chassis systems, infotainment, and software‑defined vehicles.  

With results like these, DevStudio.ai is well‑positioned to help automotive brands handle the growing pressure of delivering good quality software at speed. This, without compromising on safety or industry standards.