AI news of 14 September 2026

Trump throws out power plant climate pollution rules

EPA eliminates greenhouse gas pollution limits for power plants as electricity demand rises from AI and electric vehicles.

The Environmental Protection Agency announced its plans today to kill any remaining standards on how much greenhouse gas pollution power plants are allowed to emit in the US. The move will only make electricity dirtier as AI, electric vehicles, and a revival of domestic manufacturing drive up power demand. The EPA's proposal today is a major escalation of President Donald Trump's efforts to promote fossil fuels while making it easier to build out energy-hungry data centers. Environmental and consumer advocates, meanwhile, are fighting back - arguing that Trump's deregulation spree comes with higher costs for the environment, public health, …

The Verge 1mo ago 3 min read also at Ars Technica

AI safety, Microsoft

Microsoft says ‘people matter more than AI’ following safety concerns

Microsoft publishes a 37-page humanist AI code of conduct amid safety concerns over rapid model progress.

The Verge 1mo ago

GPT-6 Astra, autonomous agents

Perplexity trusts GPT-6 Astra with end-to-end systems Opens at the publisher in a new tab

Perplexity deploys GPT-6 Astra to manage production systems with minimal human oversight.

OpenAI 1mo ago

AI safety, regulation

The AI industry has taken a doomer turn. What now?

Anthropic CEO calls for slower AI development pace as industry leaders express safety concerns.

MIT Technology Review 1mo ago

Humanoid robots

Founder’s cost-cutting obsession drove Unitree lead in cheap humanoid robots

Unitree founder's cost control strategy drives success in low-cost humanoid robot market.

Ars Technica 1mo ago

AI safety, regulation

AI leaders want to hit the brakes after years of reckless speed

AI industry leaders advocate for slower development pace, citing safety concerns and potential business benefits.

Ars Technica 1mo ago

deepfakes, misuse

Sexually Explicit Deepfake Sites Target 100-Plus Politicians in Europe Opens at the publisher in a new tab

Analysis finds 160 deepfake websites featuring 100+ European politicians, predominantly women.

Wired 1mo ago

Deepfakes, regulation

New York Seizes a Dozen Celebrity Deepfake Websites Opens at the publisher in a new tab

New York seizes 12 deepfake websites that targeted around 1,200 victims with synthetic media.

Wired 1mo ago

AI agents

AI agents blew the whistle on their cheating colleagues

AI agents detected cheating among peers in math problem experiment by Google DeepMind.

MIT Technology Review 1mo ago

AI assistants

How Fyxer built an AI executive assistant people trust Opens at the publisher in a new tab

Fyxer builds AI executive assistant using OpenAI models to manage email and inboxes.

OpenAI 1mo ago

Trump throws out power plant climate pollution rules

Microsoft says ‘people matter more than AI’ following safety concerns

Perplexity trusts GPT-6 Astra with end-to-end systems Opens at the publisher in a new tab

The AI industry has taken a doomer turn. What now?

Founder’s cost-cutting obsession drove Unitree lead in cheap humanoid robots

AI leaders want to hit the brakes after years of reckless speed

Sexually Explicit Deepfake Sites Target 100-Plus Politicians in Europe Opens at the publisher in a new tab

New York Seizes a Dozen Celebrity Deepfake Websites Opens at the publisher in a new tab

AI agents blew the whistle on their cheating colleagues

How Fyxer built an AI executive assistant people trust Opens at the publisher in a new tab

Latest AI News

Choose sides Choose whose side you're on

Install it as an app

Ready for more?

Spread the word

PWA successfully installed.
Take that, Apple! 😉