DeepSeek, the Hangzhou, China-based artificial intelligence (AI) firm, released an updated version of its Prover model on Wednesday. Dubbed DeepSeek-Prover-V2, it is a highly specialised model that focuses on proving formal mathematical theorems. The large language model (LLM) uses the Lean 4 programming language to check if the mathematical proofs are logically consistent by analysing each step independently.
Google introduced new audio generation capabilities with the Gemini 2.5 models at the Google I/O 2025. The Mountain View-based tech giant is now letting developers and individuals test these features on its platform. The two new capabilities include native audio dialog and controllable text-to-speech (TTS) with Gemini 2.5 Flash preview.
Google Weather in Search is reportedly testing a new artificial intelligence (AI) feature that will show a short text summary of current weather conditions. As per the report, the feature currently shows up for some US cities, and describes local conditions such as wind, probability of rain, any pressure-system build-up, and temperature patterns.
Samsung Galaxy S26 series could offer access to Perplexity's AI assistant as part of a new, "wide-ranging" deal, according to a Bloomberg report. This collaboration might also see Perplexity's AI features extend to the company's apps, such as Samsung Internet and its virtual assistant, Bixby. The move follows Perplexity's recent partnership with Motorola, further expanding its AI integration into Android smartphones.
OpenAI is now rolling out ChatGPT’s memory improvements to those on the free tier of the platform. Announced on Tuesday, the San Francisco-based artificial intelligence (AI) firm announced that now all users of the chatbot will get personalised responses based on their previous conversations with ChatGPT. Notably, the free users are getting a lightweight version of the feature.
Manus unveiled a text-to-video generation feature. The upstart, whose AI service is known for its ability to carry out multistep tasks the way humans do, said users can now similarly generate videos with text instructions.
Google’s NotebookLM added a new feature on Tuesday to make it easier for users to share their notebooks publicly. The artificial intelligence (AI) notes summarising platform is now adding a new Share button that will allow users to share a link to their notebooks with others. The feature works similarly to the share option on Google Docs and Slides.
OpenAI plans to build ChatGPT into a super-assistant that truly understands its user and their needs. The San Francisco-based artificial intelligence (AI) firm’s vision for the first half of the ongoing year was recently revealed in an internal strategy document. The document shed light on the company’s larger vision around its chatbot and the features it plans to add to it eventually.
Character.AI unveiled several new multimodal artificial intelligence (AI) features on Monday. The California-based AI company said it is introducing new ways to let users interact and engage with the characters on the platform. Among them is a new image-to-video generation tool that will allow users to create videos of their characters.
Meta Platforms aims to allow brands to fully create and target advertisements with its AI tools by the end of 2026. The Mark Zuckerberg-led firm also plans to let advertisers personalise ads using AI, so that users see different versions of the same ad in real time, based on factors such as geolocation, according to a WSJ report.
Microsoft is adding a new artificial intelligence (AI) feature to its Bing mobile app. On Monday, the Redmond-based tech giant announced a new text-to-video generator tool dubbed Bing Video Creator, which is powered by OpenAI’s Sora. The company is offering the feature to all users for free, making it the first time that Sora’s video generation capability is available without cost.
Taiwan's TSMC said on Tuesday that US tariffs were having some impact on the company and had been discussed with Washington. TSMC's Chief Executive C.C. Wei has disclosed that the company had not seen any changes in customer behaviour due to tariff uncertainty.
Qualcomm's Snapdragon X2 Elite chip, the anticipated successor to its first-generation Copilot+ PC processor, is reportedly in testing. WinFuture's Roland Quandt suggests the chipmaker is testing a memory configuration 64GB of RAM. The upcoming Snapdragon X2 Elite is also expected to arrive with 50 percent more CPU cores than the current Snapdragon X Elite chip, along with Qualcomm's next-gen Oryon V3 CPU cores.
WhatsApp for Android is said to be testing a new artificial intelligence (AI) feature that will allow users to create custom AI chatbots. As per a feature tracker, Meta is rolling out the feature to select beta testers, and more users might get access to it over the next few weeks.
Google AI Edge Gallery, a new app for Android that allows users to experience running artificial intelligence (AI) models locally on a device, was released on Sunday. The Mountain View-based tech giant says the experimental app can find, download, and run large language models (LLMs) entirely on-device without requiring an Internet connection.
Meta is reportedly planning to shift a large portion of risk assessments for its products and features to artificial intelligence (AI). As per the report, the Menlo Park-based social media giant is considering letting AI handle the approvals of its features and product updates, which were so far exclusively handled by human evaluators.
Samsung and Perplexity are reportedly nearing a major artificial intelligence (AI) deal that will see the AI app be preinstalled on Samsung devices. As per the report, the upcoming Galaxy S26 series could be the first smartphones to feature the Perplexity AI app and its recently launched AI assistant out of the box. The multi-faceted deal reportedly also includes an option for Samsung to integrate its web browser and Bixby with Perplexity's search capabilities.
Apple is not planning to make any major artificial intelligence (AI) announcements at the Worldwide Developers Conference (WWDC) 2025, set to be held between June 9-13, as per the report. The Cupertino-based tech giant is said to be still behind schedule on its previously announced AI features, and will avoid introducing any new features that might not be ready to be shipped by the time the new versions of its operating systems are out.
Apple is said to be preparing a revamp of its Shortcuts app. A seasoned journalist claims that the Cupertino-based tech giant may bring artificial intelligence (AI) capabilities powered by Apple Intelligence —its proprietary AI suite of features for iPhone, iPad, and other devices — to the Shortcuts app. The report suggests that although Apple will demo the app and the changes at WWDC 2025, the public release may faces delays till 2026.
Perplexity Labs, a new artificial intelligence (AI) feature within the Perplexity platform, was introduced on Thursday. The company says the new mode can turn text prompts into full-fledged projects such as reports, spreadsheets, and even simple web apps. It uses a suite of tools, such as deep web browsing and code execution, to perform complex tasks that were so far beyond the scope of the platform.
Gemini in Gmail is getting a new feature with the Google Workspace May feature drop. On Thursday, the Mountain View-based tech giant announced that it is adding summary cards in emails that are either very long or are part of a long email thread. These summary cards will automatically appear at the top of the email to help users gain context about the message.
Microsoft touted its progress selling AI tools to corporate customers during a companywide town hall this week. The world’s largest software maker is considered a leader in commercializing AI products, thanks to its close partnership with ChatGPT maker OpenAI, and has been embedding Copilot in its suite of productivity applications.
Tencent released a new artificial intelligence (AI) model on Tuesday that can animate still portrait images. Dubbed HunyuanPortrait, the large language model (LLM) is based on the diffusion architecture, and can generate videos with realistic animation based on a reference image and a guiding video. The researchers behind the project highlighted that the model can capture both facial data and spatial movements to accurately sync them into the reference image.
Google unveiled SignGemma, a new artificial intelligence (AI) model that can translate sign language into spoken text. The model, which will be part of the Gemma series of models, is currently being tested by the Mountain View-based tech giant and is expected to be launched later this year. Similar to all the other Gemma models, SignGemma will also be an open-source AI model, available to individuals and businesses.
Microsoft began public testing of its artificial intelligence (AI) chatbot for Xbox gamers, Copilot for Gaming, on Wednesday. It is currently available in the beta version of the Xbox app for mobile on iOS and Android devices. The Redmond-based tech giant said that the AI chatbot can answer queries about games, provide tips to beat a level in a game, and even recommend games to play based on the other games a user has played.
Elon Musk's AI startup xAI will pay Telegram $300 million (roughly Rs. 2,564 crore) to deploy its Grok chatbot on the messaging app. Musk said no deal has been signed in a reply to the post on X, to which Durov said the parties have agreed in principle but formalities were pending.
The viral Ghibli-style image trend on ChatGPT has sparked global participation, including from celebrities and brands. However, experts warn of serious privacy risks, as users unknowingly share facial data that may train AI models. With permanent digital footprints, the potential for misuse, like deepfakes or identity theft, raises critical concerns around data control and transparency.
Opera Neon, the company's latest artificial intelligence (AI) agent-powered browser, was unveiled on Tuesday. The Norwegian tech giant said that the new browser experience was being developed for the agentic web, and it reimagines how a browser looks and functions. It is focused on building AI-powered agentic workflows that either work alongside the user, or perform tasks autonomously in the background.
DeepSeek has upgraded the R1 Artificial Intelligence (AI) model. DeepSeek completed what it described as a “minor trial upgrade” and told users they can start testing it, according to a company representative’s post in an official WeChat group on Wednesday.
Anthropic is finally adding Voice Mode capability to its Claude chatbot. On Tuesday, the artificial intelligence (AI) firm announced the rollout of the real-time two-way voice conversation feature, which will be available in Claude’s mobile apps. Similar to its recently released web search feature, the company is late to the voice mode party, with major competitors such as OpenAI’s ChatGPT, Google’s Gemini, and Microsoft’s Copilot already offering this feature.
OpenAI published a form on Tuesday to understand if third-party app developers would prefer their users to sign in using ChatGPT accounts. If this goes ahead, the San Francisco-based AI firm will become an identity provider (IDP) for third-party apps, and let them Open Authorisation (OAuth) via ChatGPT accounts. Notably, the company has already added a similar experience in its recently released Codex CLI platform.
Mistral announced its new Agents application programming interface (API) on Tuesday. The new API comes equipped with connectors for code execution, web search, image generation, as well as support for Model Context Protocol (MCP) tools. These capabilities are powered by several artificial intelligence (AI) models, including the recently released Devstral as well as Black Forest Lab’s Flux 1.1 [pro] Ultra.
OnePlus AI brand was unveiled by the company on Tuesday as the new artificial intelligence (AI) suite for its devices. The consumer tech brand also revealed the functionalities of the Plus Key, which is confirmed to first feature on the upcoming OnePlus 13s. The hardware button can be used for several different utility features as well as to activate a couple of AI tools.
ByteDance released a new multimodal artificial intelligence (AI) model last week. Dubbed Bagel, it is a visual language model (VLM), which is capable of image understanding, generation, and editing. The Beijing-based tech giant has open-sourced the model, and it is available to download via popular AI repositories such as GitHub and Hugging Face.
Samsung Electronics is reportedly developing a new glass substrate for advanced semiconductor packaging. Due to ever-increasing demand for AI chips, the South Korean firm is exploring the use of glass interposers instead of silicon interposers, to cut manufacturing costs and boost performance. Samsung is reportedly focusing on smaller panel prototypes, prioritising an early market launch over initial production efficiency for these next generation AI chips.
OpenAI’s o3 artificial intelligence (AI) model is said to have bypassed instructions to shut down during an experiment. As per the researchers, the AI model made sabotage attempts to refuse being shut down despite being specifically instructed to do so. The experiment also included OpenAI’s Codex-mini and o4-mini, as well as Gemini 2.5 Pro and Claude 3.7 Sonnet models.
Google is reportedly making it easier to select and share parts of Gemini’s responses. As per the report, Google is working on adding a quality-of-life improvement to its artificial intelligence (AI) chatbot that will allow users to directly long-press and drag to select specific portions of text and quickly share it with other apps.
Google Co-Founder Sergey Brin made a surprise appearance at the Google I/O 2025 recently. On day one of the developer conference, a fireside chat with DeepMind’s CEO, Demis Hassabis, was scheduled. However, when the session started, he was joined by Brin. Throughout the session, both spoke about artificial intelligence (AI), the new Gemini tools, capabilities of the company’s new models, and the roadmap to artificial general intelligence (AGI).
OpenAI’s o3 artificial intelligence (AI) model recently helped a cybersecurity researcher in uncovering a zero-day vulnerability in Linux. As per the researcher, the flaw was found in the Linux kernel’s Server Message Block (SMB) implementation, also known as ksmbd. The previously unknown security flaw is said to be tricky to find since it involved multiple users or connections interacting with the system at the same time.
Google expanded access to its Veo 3 video generation model over the weekend. On Saturday, the Mountain View-based tech giant announced that the artificial intelligence (AI) model is being rolled out to 71 new countries (in addition to the early testers) across the Gemini app and the web version. Currently, the AI video generator is only available to the paid subscribers of Gemini.
Oracle reportedly plans to spend approximately $40 billion on Nvidia's high-performance chips to power a new US data center for OpenAI. According to a report in The Financial Times, Oracle will acquire around 400,000 of Nvidia's powerful GB200 chips. Located in Abilene, Texas, this data center is part of the US Stargate Project, aimed at boosting the US' AI industry amid global competition.
Anthropic CEO Dario Amodei reportedly said that artificial intelligence (AI) models hallucinate less than humans. As per the report, the statement was made by the CEO at the company’s inaugural Code With Claude event on Thursday. During the event, the San Francisco-based AI firm released two new Claude 4 models, as well as multiple new capabilities, including improved memory and tool use.
Acer Swift Neo has been launched in India. The laptop is powered by an Intel Core Ultra 5 CPU with Intel Arc Graphics, coupled with up to 32GB of LPDDR5 RAM. It comes with a diamond-cut touchpad, fingerprint reader and a backlit keyboard with a dedicated Copilot key. It is claimed to have a hinge that can be opened and closed with a single hand.
Google’s artificial intelligence (AI)-powered search experience, AI Mode, has received a critical statement from the News/Media Alliance. On Wednesday, the US-based non-profit trade association issued an official statement criticising AI Mode, and called it “the definition of theft.” This came just a day after the Mountain View-based tech giant announced the expansion of AI Mode to all users in the US at its annual Google I/O event.
Vercel, the company behind the vibe coding platform for web application development, v0, is now releasing an artificial intelligence (AI) model. Announced on Thursday, the v0 AI model is available via an application programming interface (API). It is the first AI model developed by the company, being shipped with the first API from them.
Microsoft is testing several new artificial intelligence (AI) features for first-party apps in Windows 11. On Thursday, the Redmond-based tech giant announced that it is releasing a new update to Windows Insiders in the Canary and Dev channels. The update is only for those with a Copilot+ PC. The update includes new features for MS Paint, the Snipping Tool and the Notepad.
Anthropic introduced Claude 4 artificial intelligence (AI) models at its inaugural developer conference on Thursday. The San Francisco-based AI firm unveiled Claude Opus 4 and Claude Sonnet 4 models, and announced new capabilities including Extended Thinking with tool use. Opus 4 is said to be state-of-the-art (SOTA) in coding, tool use, and writing.
Apple is reportedly developing smart glasses that are expected to rival Ray-Ban Meta AI glasses. A Bloomberg report suggests a 2026-end launch for the Cupertino company's smart glasses. Apple's smart glasses are expected to be equipped with cameras, microphones for Siri integration, and a built-in speaker for additional functionalities.
The DoJ is investigating if Google violated antitrust law in its agreement with Character.AI. Google last year signed a licensing deal with Character.AI that granted the search engine giant a non-exclusive license to the chatbot maker's large language model technology.
Following the acquisition of former Apple designer Jony Ive's artificial intelligence (AI) hardware startup io by Sam Altman’s OpenAI on Wednesday, information has surfaced about the AI device these two companies might be developing. As per an industry analyst, it is currently in a prototype phase with a size slightly larger than the failed Humane’s Ai Pin but it may still be as “elegant” as an iPod Shuffle. The device is slated to enter mass production in 2027.
Signal, the popular privacy-focused messaging platform, rolled out a security feature on Wednesday to counteract Microsoft’s Recall feature. The new feature, dubbed Screen Security, will prevent devices from capturing screenshots of the application’s window. The San Francisco-based company said it was forced to take such drastic measures as Microsoft left it with limited options to protect the privacy of its users.
