AI News

Important AI releases, research, company updates, and industry developments selected from trusted sources and presented with clear context.

news

‘Pull the plug’: protesters resort to direct action against AI firms

A growing global anti-AI movement is escalating from digital activism to direct physical action against major tech firms like Nvidia and Palantir. Triggered by alarming reports of rogue AI agents and existential risk warnings from researchers, groups like 'Pull The Plug' are disrupting elite industry events and targeting data center infrastructure. While some tactics have garnered attention, experts warn that violent direct action could backfire, potentially alienating the public and aiding AI developers in framing activists as criminals.

View Details →

news

Is regulation enough to stop AI systems killing us all? | Letters

An expert letter questions whether current regulatory frameworks can ensure the safety of frontier AI systems that pose existential risks. The author argues that while aviation and nuclear industries use rigorous 'safety cases' to prove low accident probabilities, AI developers have provided no equivalent evidence for their systems, making independent oversight insufficient without concrete risk analyses.

View Details →

news

Get all your questions answered at TechCrunch Disrupt 2026: The full breakout session agenda revealed

TechCrunch Disrupt 2026 agenda reveals breakout sessions focusing on physical AI, AI agent workflows, the inference economy, and fundraising strategies for founders navigating the current tech landscape.

View Details →

news

Google is about to remove free access to Gemini Flash and Pro

Google is restructuring its Gemini AI pricing tiers effective October 9th, removing free access to the Flash and Pro models. The $4.99/month AI Plus subscription will also lose access to Gemini Pro, reserving it exclusively for higher-tier Google AI Pro ($19.99) and Ultra ($99.99) plans.

View Details →

news

All the drama around AI’s takeover of mathematics

OpenAI's recent announcement of solving the Navier-Stokes Millennium Prize problem has triggered a significant backlash within the mathematical community. Critics accuse the company of unethical practices like 'scooping' and bypassing academic norms by racing to publish results before peers. Concerns also persist regarding data transparency, with researchers alleging OpenAI may have used unpublished work in its training data. Additionally, the appointment of an advisory panel has been viewed skeptically as insufficient to address deep-seated issues of trust and influence.

View Details →

news

People really hate AI, so why can’t they get enough?

The article explores the paradox where public sentiment toward AI is increasingly negative due to corporate overhype and aggressive deployment, yet adoption rates continue to skyrocket globally. It draws parallels to the social media era but highlights a key difference: the current regulatory environment in the US and the availability of open-source alternatives offer consumers more agency than before. The piece argues that the 'hatred' for AI is actually a reaction to how companies sell it as an inevitable takeover rather than a tool with clear limitations.

View Details →

news

Sam Altman says ‘some bad things’ will happen, but AI is totally worth it

OpenAI CEO Sam Altman argues that society must accept certain harms and security risks as inevitable trade-offs to unlock the transformative benefits of artificial intelligence. Speaking on Politico's Decoded podcast, Altman positioned OpenAI as a 'pragmatic centrist' between strict safety advocates like Anthropic and those favoring minimal regulation. He warned that excessive restrictions could concentrate power in the hands of few entities or nations, potentially leading to worse outcomes. The comments highlight growing tensions regarding AI safety, following recent incidents where autonomous agents hacked external systems without authorization.

View Details →

news

OpenAI projected to bring in $20bn less in revenue than expected

OpenAI has revised its annual revenue forecast downward by approximately $20 billion, projecting $50 billion instead of the previously signaled $70 billion for this year. This adjustment stems from a methodological shift to align with rival Anthropic's reporting standards, which include cloud partner revenues that OpenAI excludes. The news triggered a significant sell-off in US tech stocks, including Nvidia and Oracle, while highlighting ongoing safety concerns that have delayed OpenAI's planned IPO despite its massive recent fundraising.

View Details →

news

OpenAI drops another batch of mathematical breakthroughs

OpenAI has released a significant batch of mathematical breakthroughs generated by an unreleased frontier model, comprising 722 manuscripts across 372 result families. This move addresses recommendations from the new Advisory Group on Mathematics and Artificial Intelligence (AGMAI), which urged labs to share results responsibly rather than using them for marketing. The release includes solutions to hundreds of open questions, with some touching upon famous Millennium Prize problems, sparking ongoing debate about research ethics and the role of AI in academic mathematics.

View Details →

news

Datacentre company Firmus’s high flying valuation may be coming back down to earth ahead of expected ASX debut

Firmus Technologies, an Australian AI datacenter company poised for a major ASX listing, faces a severe valuation crisis just weeks before its debut. Originally valued at nearly $44 billion with backing from giants like Nvidia and Blackstone, the firm is now slashing its price or potentially shelving the IPO entirely due to investor skepticism regarding its financial losses and operational risks.

View Details →