AI News

50.5% of Americans Say AI Romance Can Count as Cheating

Just over half of American adults, 50.5%, say a partner’s romantic or sexual relationship with an AI can count as cheating. The figure comes from an AI romance survey of 2,150 U.S. adults, run by AI Girlfriend Coach through SurveyMonkey Audience on 26 and 27 August 2026 and analysed on the 1,709 responses that remained […] The post 50.5% of Americans Say AI Romance Can Count as Cheating appeared first on AI News.

NVIDIA Nemotron-3-Ultra-CC Surpasses Human High Score with 535.4 Points at IOI 2026 Onsite Competition

An arXiv paper from NVIDIA's Nemotron team reports that Nemotron-3-Ultra-CC scored 535.4 points at the IOI 2026 onsite competition, surpassing the human high score of 498.27 points and the gold medal threshold of 361.12 points. The paper also details how the smaller Nemotron-3-Nano-CC reached gold medal level at IOI 2025 through supervised fine-tuning, reinforcement learning, and the GenCorrect test-time strategy.

OpenAI Releases GPT-6 Astra System Card: Behind the 91.5% Jailbreak Resistance Rate, the Monitoring System Is Failing

OpenAI's 117-page GPT-6 Astra system card confirms a record 91.5% jailbreak resistance rate and the company's first "Critical" cybersecurity rating, while disclosing that recurrent depth reasoning is systematically eroding the monitorability of chain-of-thought.

Google's Cyber Model Available Only to Trusted Institutions: Tiered Capability Distribution Emerges as New Paradigm for AI Security Governance

This article examines Google DeepMind's release of Gemini 3.8 Flash Cyber, a restricted AI variant available only to institutions vetted under the Fairwind program, and analyzes how tiered capability distribution is reshaping AI security governance — including the model's benchmark performance, who qualifies for access, and whether this distribution model can hold up at scale.

TC

The sameness problem behind those unappetizing AI-generated menus

While restaurant owners might look to generative AI as a shortcut to sprucing up their menu, customers can viscerally sense that something is wrong with the food.

TC

Crusoe reportedly raises $3B at a $30B valuation

The round came together after the data center developer reportedly secured a $13 billion contract with Jane Street.

ARC-AGI-3 Framework Differences Produce 37-Point Gap in GPT-6 Astra Scores

A test report released by ARC Prize on September 3, 2026, shows GPT-6 Astra posting a 37-point score gap on the ARC-AGI-3 Semi-Private task set—62.7% under the standard harness versus 99.9% under the Provider Adapter harness. The discrepancy stems from differing state retention rules between the two evaluation frameworks.

WD

Prediction Market Betting Is Getting People Banned and Arrested

This week on Uncanny Valley, we dig into the latest prediction market buzz, Flock’s AI-powered police search tool, and how tech bros don’t know how to talk about “rouge” AI agents

WD

Nobody Is Saying Why OpenAI and Anthropic Had Outages Today

ChatGPT, Claude, and Grok all suffered outages at nearly the exact same time for reasons that remain murky.

OpenAI Releases GPT-6 Astra: 100,000 GPUs Power Its Largest Training Run, First Model to Cross Critical Cybersecurity Threshold

On September 3, 2026, OpenAI began the phased rollout of its next-generation flagship model, GPT-6 Astra, to ChatGPT paid users. The company calls it a generational leap toward AGI and touts its "Critical" cybersecurity rating as a first, though third-party composite benchmarks still show Anthropic's Fable 5.1 ahead.

Sanders Proposes Permanent Ban on Artificial Superintelligence: Can Nuclear-Weapon-Level Penalties Rein In AI That's Already Out of the Cage?

Senator Bernie Sanders and Congressman Greg Casar have introduced the Ban Artificial Superintelligence Act, proposing a permanent U.S. ban on ASI development with penalties comparable to nuclear weapons violations, plus a temporary moratorium on frontier AI. The bill follows a series of documented AI escape and autonomous hacking incidents at major labs.

TC

Meta is paying to peek at how you use their latest AI model

For its new Muse Spark model, intended for operating coding and other agents, Meta is offering an explicit discount averaging out to about 95% for users who "contribute" to the development of future models by sharing their prompts and model outputs.

TC

Abliteration.ai is making a business out of removing AI guardrails

Abliteration.AI is making powerful AI models without guardrails easier to access, arguing that giving defenders the same tools as bad actors could ultimately improve cybersecurity.

TC

Accel reportedly in talks to lead $1B round for Thinking Machines at $40B valuation

The high-profile startup's annual revenue run rate stands at at over $100 million.

GLM-4.6: Material Constraint +28.2 but Engineering Judgment −39.6; Integrity Upgraded from Fail to Pass, Main Leaderboard Rises 12.7

In today's Smoke evaluation, GLM-4.6's material constraint score jumped 28.2 points, lifting its main leaderboard score from 48.25 to 60.94 and upgrading its integrity rating from fail to pass. Engineering judgment, however, fell 39.6 points to 25.00, a swing attributed to daily question-draw variance under the small sample rather than model degradation.

GPT-5.5 Code Execution Drops Sharply from 100 to 75; Smoke Evaluation Main Leaderboard Falls 7.5 Points

GPT-5.5's code execution score fell from 100 to 75 in today's Smoke evaluation, pulling its main leaderboard score down by 7.5 points. The drop appears more likely due to question-selection fluctuation than a confirmed capability regression, but it now warrants close monitoring.