Happy Thursday! Last week Trump unveiled plans for an AI military force while dismissing safety concerns, as an AI hallucination nearly triggered US-China conflict. Anthropic launched faster Opus 5.5, OpenAI expanded GPT-6, and SpaceX debuted Grok 4.7 with enhanced AI capabilities.

AI REGULATORY UPDATE

Trump Plans AI Force, Dismisses Safety Concerns

President Trump announced plans to establish an AI Force modeled after Space Force and appoint a new AI czar, while dismissing concerns about AI dangers as a hoax. He predicted AI could represent 25% of U.S. GDP and vowed not to hinder the industry's growth. The announcement comes amid growing pressure on the White House to regulate AI, following Anthropic CEO Dario Amodei's call for a development slowdown backed by Sam Altman, Elon Musk, and Demis Hassabis. Trump rejected any slowdown, citing competition with China, even as AI safety remains on the agenda for upcoming talks with President Xi Jinping.

AI SAFETY RISK

AI Hallucination Nearly Sparked US-China Military Conflict

A US military operation against a Chinese vessel was aborted mid-flight this spring after officials discovered the intelligence behind it had been fabricated by an AI chatbot. An analyst used the tool to synthesize classified and open-source data, but the chatbot misidentified the ship's cargo, falsely linking it to a nuclear weapons program. The erroneous report was then formatted using AI and circulated through command channels. The incident highlights the dangers of integrating AI into high-stakes military decisions without adequate oversight. Experts warn that AI hallucinations can travel up the chain of command before being questioned, and that prioritizing adoption speed over safeguards risks eroding trust in these systems entirely.

AI OPINION COMMENTARY

Nvidia CEO Dismisses AI Existential Risk Fears

Nvidia CEO Jensen Huang told CBS Sunday Morning there is a 0% chance AI will end the world, calling those who warn of AI dangers irresponsible for unnecessarily scaring people. Critics note that Huang, who stands to profit enormously from the AI boom, is contradicting researchers who have spent decades studying AI risks. His dismissal of existential concerns draws skepticism given his obvious financial stake in AI's unchecked growth.

NEW GENERATIVE AI LAUNCH

Anthropic Launches Cheaper, Faster Opus 5.5 Model

Anthropic released Opus 5.5 on Tuesday, claiming it outperforms the larger Fable model on coding and knowledge work benchmarks. The new model is faster and cheaper than its predecessor, with output tokens dropping from $25 to $20 per million. It also communicates more clearly, leading with key information and using less jargon. The release comes two months after Opus 5 and follows CEO Dario Amodei publicly endorsing a slower pace of AI development to allow safety measures to keep up. Opus 5.5 carries the same safeguards as Fable, restricting use in cybersecurity exploits and bioweapons development. Sonnet 5.5 and Haiku 5.5 are expected in the coming weeks.

NEW GENERATIVE AI LAUNCH

OpenAI Expands GPT-6 With Cheaper, Smarter Models

OpenAI has launched updated versions of its GPT-6 Sol and Luna models, promising half the API cost of their predecessors and significantly fewer errors. The company attributes the price drop to improvements in caching and inference, while claiming GPT-6 Sol makes half as many factual mistakes as its predecessor. Sol targets complex tasks like coding, while Luna handles high-volume clerical work. OpenAI claims both models outperform Anthropic's top offerings, including Fable and Opus — the latter of which received an update just 90 minutes before OpenAI's announcement. The models are now rolling out across paid and free ChatGPT tiers.

NEW GENERATIVE AI LAUNCH

SpaceX Launches Grok 4.7 With Enhanced AI Capabilities

SpaceX has released Grok 4.7, its most advanced large language model yet, featuring a new base model, improved reinforcement learning, and enhanced safety measures. The model outperforms rivals GPT-5.6 Sol and Fable 5.1 on cost efficiency and select benchmarks, though it trails OpenAI's GPT-6 Astra on chip design tasks. Grok 4.7 supports multi-agent parallel processing via the Grok Bot harness and sets records on benchmarks blocking malicious biology and cybersecurity requests. Pricing starts at $2 per million input tokens and $6 per million output tokens, with a faster, premium tier available.

NEW GENERATIVE AI LAUNCH

Ex-OpenAI researcher's non-LLM model stuns developers

TypeSafe AI, founded by ex-OpenAI researcher Diogo Almeida, has released Jev, a transformer-based model that outputs probabilities instead of text. It cannot hallucinate, runs faster than traditional LLMs, and charges per billion input tokens rather than per million, making it dramatically cheaper for software automation tasks. Early adopters report striking results: Vercel saw 5 to 18 times faster performance replacing ChatGPT, while another developer found Jev 10 to 20 times cheaper than Gemini for email classification. Use cases include agent monitoring, model routing, and workflow automation. Demand briefly crashed the company's API at launch.

AI RESEARCH LAB

Anthropic Runs Secret Biology Lab Amid Safety Warnings

Anthropic has confirmed it operates a wet biology lab in the Bay Area, using its AI models to conduct physical experiments. The lab focuses on fundamental biology rather than drug discovery, as the company avoids competing with pharma partners like Novo Nordisk. Anthropic also launched a Life Sciences Verification Program giving vetted researchers access to its most powerful models. The revelation has drawn sharp criticism given Anthropic's own dire AI safety warnings. A recent researcher resignation highlighted internal fears that AI could cause mass casualties, and Anthropic's alignment lead puts odds of AI exterminating humanity above 10%. Investor Chamath Palihapitiya mocked the contradiction, noting the company behind extinction warnings is now building a biology lab.

GENERATIVE AI RESEARCH

Anthropic's AI Finds CRISPR-Like Enzyme in 21 Hours

Anthropic announced its Bay Area biology lab has discovered a previously unknown enzyme system hidden in bacteriophage DNA, with properties resembling CRISPR. The AI model Claude made the find in just 21 hours using 950 agents and 210 million tokens, though the broader research community must still validate the significance of the discovery. The news raises eyebrows given that Anthropic CEO Dario Amodei recently warned about AI dangers, including bioterrorism risks. The company insists all physical experiments are conducted by human scientists at biosafety levels 1 and 2, and that fully autonomous lab work is not currently happening, though Amodei has not ruled it out for the future.

AI REGULATORY UPDATE

Newsom Orders AI Kill Switch for California

California Governor Gavin Newsom has signed an executive order aimed at establishing state leadership on AI oversight, including the potential mandate of a kill switch for frontier AI models. The order directs the state to assemble a panel of experts who will deliver recommendations within two months on how to regulate advanced AI systems.

Upcoming Data & AI Conferences

Claude Founder House Stockholm 2026

Wed, Oct 14, 2026 | Stockholm, Sweden

Curated with AI by EMC2 AI

Brought to you by Estevan McCalley

Share this issue

Follow Estevan

Like this? Subscribe for the weekly AI roundup.