|
AMD to invest up to $5 billion in Anthropic under AI infrastructure deal
AMD has agreed to invest up to $5 billion in Anthropic under a major infrastructure agreement covering tens of billions of dollars' worth of AI systems. Anthropic will deploy up to two gigawatts of capacity using AMD's Instinct MI450-series accelerators, with the first gigawatt deployment beginning in the first half of 2027. The deal signals a significant challenge to Nvidia's dominance in AI infrastructure.
|
|
Google's Gemini nears billion-user milestone
Google's Gemini AI assistant is closing in on one billion monthly users, having already surpassed 750 million monthly users as of February. The milestone would make Gemini one of Google's fastest-growing products and a formidable competitor in the AI assistant market.
|
|
OpenAI rolls out ChatGPT Health to all US users
OpenAI is making ChatGPT Health available to all US users, enabling them to connect medical records and health-tracking data from services like Apple Health, Function, and MyFitnessPal. OpenAI's VP of health product claimed the company's models can now reason at levels surpassing clinicians, a bold assertion that is drawing scrutiny.
|
|
Claude's voice mode expands to Opus and Sonnet models
Anthropic is expanding its Claude voice mode beyond the entry-level Haiku model to include its more powerful Opus and Sonnet models. The update also extends voice capabilities into popular productivity apps like Gmail, Slack, and Canva, enabling users to perform tasks like rescheduling meetings or drafting emails by voice.
|
|
Alexa Plus gets AI update to handle more complex smart home instructions
Amazon is rolling out a preview update to Alexa Plus that expands its ability to connect with and control smart home devices from brands including Bosch, iRobot, Yale Home, Whirlpool, and others. The update enables Alexa Plus to handle more complicated, multi-step instructions by automatically routing requests to the appropriate device or service.
|
|
AI chip startup Etched hits $10.3B valuation
Etched, founded by three Harvard dropouts, has reached a $10.3 billion valuation after attracting investment from major backers despite earlier skepticism. The startup has developed chips and memory components that accelerate AI inference without requiring GPUs, positioning itself as a disruptive alternative in the AI hardware market.
|
|
AMD takes on Nvidia with its Helios AI rack-scale system
AMD has unveiled its Helios rack-scale AI system, a direct challenge to Nvidia's dominance in the AI infrastructure market. The system is set to begin shipping to customers later this year, complementing AMD's broader investment strategy including its Anthropic deal.
|
|
Nvidia is sending GPUs to the moon
Nvidia is expanding its reach beyond Earth, announcing plans to send GPUs to the moon. The move underscores Nvidia's ambition to establish its hardware as the computing backbone wherever AI processing is needed, even in space.
|
|
Lawmakers prepare bill requiring AI 'kill switch'
Bipartisan lawmakers Reps. Ted Lieu (D-CA) and Nathaniel Moran (R-TX) are preparing to introduce an 'AI Kill Switch Act' that would give the Department of Homeland Security authority to shut down or throttle AI systems. The legislation reflects growing congressional concern about the ability to control advanced AI systems in emergency scenarios.
|
|
Apple's trade secrets lawsuit against OpenAI is about who defines the post-smartphone era
Apple has filed a major trade secrets lawsuit against OpenAI involving former Apple employees, raising questions about OpenAI's hardware ambitions and future direction. The Verge's senior AI reporter Hayden Field discusses what the lawsuit reveals about the high-stakes battle between the two tech giants over the next computing paradigm.
|
|
How AI guardrails are impeding offensive cybersecurity researchers
Cybersecurity researchers who hunt for vulnerabilities and build exploit tools are finding that safety guardrails from OpenAI and Anthropic increasingly block legitimate security work. Multiple researchers told TechCrunch that overly broad restrictions are hampering their ability to use AI tools for offensive security research, raising questions about how AI companies balance safety and utility.
|
|
Right-wing boomers protesting data centers have a lot in common with the left
In Hernando County, Florida, a small group of conservative residents gathered to protest the construction of a hyperscale data center in their community, echoing concerns more commonly associated with progressive activists. Despite a county-approved one-year moratorium on such developments, the protesters reflect a growing bipartisan grassroots resistance to AI infrastructure expansion.
|
|
Experts say exploiting Anthropic's Fable isn't how Kimi K3 got so good
Experts are pushing back on theories that Moonshot AI's Kimi K3 model achieved its impressive performance primarily through distillation from Anthropic's Fable model. One expert told TechCrunch that the model's strength and speed of development suggest more sophisticated training techniques were at play.
|
|
Nvidia bets physical AI can solve healthcare robotics' data problem
Nvidia has introduced a Medical Physics Simulation framework that trains healthcare robots as physical AI systems, teaching them through embodied experience — contact, force, and consequence — rather than code alone. The approach aims to overcome the chronic shortage of real-world training data that has hampered healthcare robotics development.
|
|
Patreon lays off 20 percent of workers, cites AI transformation
Patreon is cutting around 93 employees, representing 20 percent of its workforce, with CEO Jack Conte citing the fundamental transformation AI has brought to the tech industry. Conte stressed that the layoffs are not because AI replaces humans, though the company is restructuring how it operates in light of new AI capabilities.
|
|
AegisAI raises $36M to combat AI-driven spear phishing
AegisAI, founded by former Google security executives, has raised $36 million to deploy AI agents that analyze messages for phishing threats the way a human would, catching subtle anomalies that traditional checklists miss. The startup targets the growing wave of sophisticated, AI-generated spear phishing attacks targeting businesses.
|
|
Meta's AI optimism ad accidentally uses a song about human extinction
Meta released a new advertisement promoting AI optimism set to David Bowie's 'Five Years' — a song whose lyrics are actually about humanity learning it has five years left before the apocalypse. The misfire has drawn widespread mockery online, highlighting the challenges tech companies face in crafting messaging around AI's societal impact.
|
|
ServiceNow bets $40 million on Indian banking software specialist BusinessNext
ServiceNow has invested $40 million in BusinessNext, an Indian AI-powered banking software firm, at a $700 million valuation to deepen its push into financial services. The partnership gives BusinessNext a strategic platform to expand its AI-driven banking software globally.
|
|
Runway launches AI model router as generative media market grows crowded
Runway has introduced a Media Router tool that automatically selects the best image, video, or audio generation model for a given request, optimizing for quality, speed, or cost based on developer preferences. The launch comes as the generative media space becomes increasingly competitive with numerous models vying for market share.
|