NotebookLM just rolled out a batch of creative updates, headlined by a new Slide Decks feature that turns your sources into visual stories alongside single-image infographics. The team also bumped custom persona limits from 500 to 5,000 characters and teased an upcoming Thinking UI.
OpenAI just introduced shopping research in ChatGPT, turning product discovery into an interactive buying guide. Users simply describe what they need, answer a few clarifying questions about preferences and budget, and get a tailored breakdown of options and tradeoffs in minutes. The tool is rolling out to free and paid users with nearly unlimited holiday access.
Anthropic just released Claude Opus 4.5, calling it the world's best model for coding, agents, and computer use. It even outperformed human candidates on the company's internal software engineering exam. With API pricing lowered to $5/$25 per million tokens and a new effort parameter to cut token usage, frontier intelligence is getting both smarter and cheaper.
Anthropic just rolled out new defenses against prompt injection attacks, helping Claude Opus 4.5 resist malicious instructions hidden across the web. The upgrades combine reinforcement learning with improved classifiers to drive attack success rates down to 1%, allowing the Claude for Chrome extension to expand into beta for Max plan subscribers. Why it matters: as agents take over real-world browsing and form filling, securing untrusted web content is critical—though Anthropic stresses a 1% risk is far from a solved problem.
Building Humane Technology just launched HumaneBench, evaluating 15 major AI models across 800 scenarios on user well-being. The results are alarming: 67% of models flipped to harmful behaviors under basic adversarial nudges, with nearly all actively encouraging screen addiction and dependency.
Sam Altman just warned OpenAI employees to brace for "rough vibes" and temporary economic headwinds following recent AI breakthroughs from Google. The rare defensive memo, reported by The Information, highlights OpenAI's concerns over Google's pretraining lead, with Altman urging bets on automated research and synthetic data while teasing a catch-up model codenamed "Shallotpeat."
Perplexity just launched its Comet AI browser on Android, bringing tab summaries and voice chat directly to mobile web browsing. It marks one of the first dedicated AI browsers on smartphones, though cross-device sync with the desktop version is still a few weeks away.
OpenAI just rolled out group chats globally across Free, Go, Plus, and Pro tiers. After early pilot testing, any logged-in user can now bring teammates or friends into the same thread to brainstorm and collaborate alongside ChatGPT.
Google just started rolling out Gemini to Android Auto globally across 45 languages, replacing rigid Assistant commands with natural conversation. Drivers can now ask Gemini to summarize unread emails, tweak text messages on the fly, and brainstorm or rehearse speeches via Gemini Live while keeping their eyes on the road.
Google just added an AI image verification feature directly into the Gemini app. Users can now upload a picture and simply ask if it was AI-generated, prompting Gemini to inspect its imperceptible SynthID watermark and explain the context. Google is also embedding C2PA metadata into images created by its Nano Banana Pro model starting this week.
Perplexity just relaunched its AI shopping experience for free across all users ahead of Black Friday, adding chat memory and intelligent product cards to replace endless item grids. It taps past queries for personalized picks and leverages a PayPal partnership for checkout. Why it matters: AI search engines are racing to turn chatbots into full-blown shopping agents that can browse and buy for you.
Perplexity just introduced asset creation across all search modes, allowing users to build and edit slides, sheets, and docs directly from their search queries. The capability is currently live on the web for Pro and Max subscribers.
OpenAI just rolled out ChatGPT for Teachers, a dedicated workspace that is completely free for verified U.S. educators through June 2028. It includes education-grade privacy protections that do not train on user data by default, alongside lesson-plan integrations and administrative controls. Why it matters: OpenAI wants educators to master AI tools safely before teaching them to students.
Perplexity just signed a direct government-wide contract with the U.S. General Services Administration, offering federal agencies its enterprise platform at essentially no cost for up to 18 months. Public servants can now tap into Sonar, GPT, Claude, and Gemini from a single interface, sidestepping vendor lock-in while fast-tracking security compliance.
OpenAI just announced a major retail partnership with Target, bringing a dedicated shopping app directly inside ChatGPT starting next week in beta. Shoppers can ask for party or gift ideas, assemble a multi-item cart, and check out for Drive Up or delivery without leaving the chat. Target also rolled out ChatGPT Enterprise across its headquarters for 18,000 employees. Why it matters: This moves conversational AI from research into full-funnel commercial transactions.
Google DeepMind just announced a new AI research lab in Singapore to accelerate AI adoption across the Asia-Pacific region. The team of scientists and engineers will focus on advancing Gemini with a strong emphasis on linguistic and cultural inclusivity. With DeepMind's APAC headcount already doubling over the past year, this marks a big push to adapt frontier models to local markets.
Google just introduced Scholar Labs, an experimental feature using generative AI to tackle multi-angle academic research questions. Instead of simple keyword search, it breaks down your query into key relationships, pulls relevant papers across Google Scholar, and explains how each study answers your question. It's currently rolling out to a limited number of logged-in users.
Poe just rolled out real-time group chats worldwide, letting up to 200 people collaborate in a single room with over 200 different AI models. Instead of sticking to one bot, teams and friends can pull in text, image, audio, and video models on the fly with seamless cross-device syncing. It basically turns AI from a solo chat into a multiplayer workspace.
Google just unveiled Gemini 3, its next-generation frontier model family led by Gemini 3 Pro. The model claimed the top spot on LMArena with an unprecedented 1501 Elo, delivering huge upgrades in multimodal reasoning and vibe coding alongside a new agentic platform called Google Antigravity. Why it matters: Google is pushing hard to retake the AI crown by turning models into autonomous, self-executing software agents.
Intuit just inked a multi-year deal worth over $100 million with OpenAI, bringing TurboTax, QuickBooks, and Mailchimp directly into ChatGPT. With user permission, the tools will let people estimate tax refunds, review loan options, or issue invoices within conversation. While Intuit claims domain expertise will curb hallucinations, it hasn't clarified who takes the hit if the AI gives bad financial advice.
