|
Google announces Gemini 4 Argon, its most powerful model yet
Google has unveiled Gemini 4 Argon, its latest frontier AI model touted for excelling in complex workflows including software engineering, legal and finance work, and cybersecurity defense. Access is currently restricted to 'trusted cyber defenders' given the model's powerful capabilities. The release marks a significant step forward in Google's AI ambitions.
|
|
Meta disputes claim that Muse AI agent read user's private messages without permission
Meta has pushed back against a report that its new Muse AI agent accessed a journalist's private messages without authorization, insisting the agent cannot read Messages without explicit user permission. The controversy highlights growing concerns about privacy and the scope of permissions granted to AI agents. Meta's Muse, which can help with tasks from sending emails to online shopping, has faced scrutiny since its launch.
|
|
ElevenLabs doubles valuation to $22B in $300M employee tender offer
AI voice startup ElevenLabs has doubled its valuation to $22 billion through a $300 million employee tender offer co-led by Wellington and T. Rowe Price. The deal underscores strong investor appetite for AI audio and voice technology. ElevenLabs continues to be one of the most highly valued AI startups in the space.
|
|
Flow Engineering raises funding at $750M valuation backed by Sequoia, Valor, and Atreides
AI startup Flow Engineering, which applies AI agents to hardware design workflows, has secured new funding at a $750 million valuation from Sequoia, Valor, and Atreides. Sequoia's Roelof Botha also joined as an angel investor and board member. The deal reflects continued investor enthusiasm for AI applied to engineering and chip design.
|
|
The ugly economics of consumer AI
Frontier AI labs have grown cautious about consumer-facing AI products, and the reasons go beyond technical readiness. The economics of consumer AI remain challenging, with high costs, uncertain monetization, and fickle user retention making it a risky bet. The piece examines why building sustainable consumer AI businesses is harder than it looks.
|
|
Here's how tech leaders will self-police AI safety under Trump's deal
President Trump has announced a 'morally binding' AI safety agreement with top tech executives, officially titled the Joint Commitment on Frontier Responsibilities, which calls for industry self-regulation. AI leaders have weighed in with mixed reactions to the voluntary accord, which was shared online by presidential adviser David Sacks. Trump also walked back earlier comments calling AI safety concerns a Democratic 'hoax,' now that he has rebranded the technology as 'Super Intelligence.'
|
|
Reddit kills RSS feeds and ends public API access to fight AI bots
Reddit is shutting down RSS feed support and ending public API access as part of an ongoing effort to restrict access to its large store of user-generated content amid abuse by AI bots. The move continues a pattern of tightening data access that Reddit began in previous years. The decision is likely to affect developers and third-party services that relied on open access.
|
|
Google reportedly tests paying publishers for AI search results
Google has launched a pilot program involving around 100 publishers to compensate them for content used in its AI-powered search features, according to reports from The Information and Digiday. The program comes amid growing scrutiny over AI's impact on web traffic and the economics of online publishing. The move could signal a broader shift in how AI companies handle content licensing.
|
|
The AI Tamagotchis are coming: Meta and OpenAI bet on physical AI hardware
Meta and OpenAI are both developing dedicated AI hardware devices in an effort to break through where previous AI gadgets have failed. Both companies are reportedly testing consumer appetite with cute, companion-style software agents before committing to full hardware rollouts. The strategy mirrors a Tamagotchi-like approach to building emotional attachment to AI products.
|
|
OpenAI's 'Decisions API' is a Jev clone aimed at controlling swarming agents
OpenAI has quietly developed a 'Decisions API' modeled after Jev, designed to help manage and stop swarms of autonomous AI agents. The tool confirms the growing importance of fast, low-cost decision intelligence in agentic AI systems. The move signals OpenAI's focus on building infrastructure to govern increasingly complex agent deployments.
|
|
Destro AI bridges the gap between robots and humans
Destro AI is differentiating itself in the robotics space by focusing on human-robot collaboration rather than fully autonomous robots. The company says its edge comes from making robots and humans work seamlessly on the same page, rather than replacing human roles entirely. Its approach is helping it compete against and outperform traditional robotics companies.
|
|
DoorDash launches an AI agent you can text to order food
DoorDash has introduced a conversational AI agent that lets customers order food by simply sending a text message. The feature is designed to give DoorDash a competitive edge over rivals Uber Eats and Grubhub. The launch reflects a broader trend of food delivery platforms integrating AI to streamline ordering experiences.
|
|
Airbnb adds AI search and new social features
Airbnb has rolled out AI-powered search capabilities alongside new social features to enhance the travel booking experience. The platform is also launching additional services such as meal delivery and laundry in select locations. The updates represent Airbnb's push to expand beyond accommodation into a broader travel and lifestyle platform.
|
|
Restate lands $20M to build durable infrastructure for AI agents
Restate has raised $20 million to develop its durable execution engine, which is purpose-built to support the reliability demands of AI agent workflows. Unlike competitors, Restate built its own storage, replication, and redundancy layers rather than relying on external databases, enabling exceptional speed and lightweight performance. The funding reflects growing demand for robust infrastructure as AI agents take on more critical tasks.
|
|
Instagram adds AI 'creative assistant' to help creators improve their posts
Instagram has introduced an AI-powered 'creative assistant' within its standalone Edits app that analyzes a user's account data and offers personalized posting suggestions. The feature joins a growing list of social media platforms offering AI-driven content guidance to creators. Instagram is also eyeing features inspired by YouTube as it competes for creator audiences.
|
|
Grokipedia gets a 'newly refreshed' design in v0.3 update
Grokipedia, SpaceXAI's AI-powered Wikipedia rival, has received a visual overhaul as part of its v0.3 update, including a new logo and refreshed homepage and live edits page. The update comes after the platform recently resumed accepting user edits. SpaceXAI's head of design described the result as a 'newly refreshed Grokipedia.'
|
|
Instinct's new AI product recommendations are getting a mixed reception
Instinct is rolling out human-curated product and travel recommendations to its users, but the unsolicited suggestions are rubbing some people the wrong way. Critics say the feature feels intrusive, as users did not opt in to receiving recommendations. The backlash highlights the fine line AI-powered apps must walk between helpful personalization and unwanted interruption.
|
|
Cerebras CEO Andrew Feldman to discuss AI scaling limits at TechCrunch Disrupt 2026
Cerebras Systems CEO Andrew Feldman will take the stage at TechCrunch Disrupt 2026 to address whether AI can continue to scale amid rising demand for compute, energy, and infrastructure. Feldman will outline how Cerebras is approaching hardware constraints differently from the rest of the industry. The talk will also explore what happens if today's AI hardware hits its physical limits.
|
|
3 days left to exhibit at TechCrunch Disrupt 2026
Startups have just three days left to secure an exhibitor spot at TechCrunch Disrupt 2026, with the deadline set for October 2 at 11:59 p.m. PT. The event will bring together more than 10,000 founders, investors, operators, and tech leaders. Exhibiting offers startups a key opportunity to gain visibility with a major tech audience.
|