Today's AI page · AI digest archive

AI digest · 2026-09-23

The 10 AI stories that the most sources reported on this day, each with the reports it was drawn from.

  1. Anthropic's wet lab yields AI-discovered enzyme akin to CRISPR — First major result from Claude-driven biology research, though humans remain in the loop

    Sources:[3] theverge.com[7] techcrunch.com

  2. Sanders proposes prison sentences for developing "superintelligence" — Sweeping US legislation would ban AI capable of "destruction or disempowerment of humanity."

    Sources:[4] theverge.com

  3. OpenAI and Anthropic slash model prices 40-80% amid fierce competition — Price war intensifies as Luna, Opus, and open models drive token costs down dramatically

    Sources:[28] simonwillison.net[30] latent.space[75] tomtunguz.com

  4. Google launches Gemini 3.8 Flash TTS with custom voice design from text — New text-to-speech models support 100+ languages, stage directions, and two-voice dialogue

    Sources:[13] the-decoder.com[23] deepmind.google[25] simonwillison.net

  5. ChatGPT Voice upgrades to GPT-6 Astra with email, calendar, Slack access — Mobile agentic features bring conversational AI closer to full personal assistant capabilities

    Sources:[9] techcrunch.com[12] the-decoder.com

  6. YouTube rolls out AI creator tools: script coaching, Gemini editing, live translation — Platform increasingly automates content strategy and production decisions for creators

    Sources:[5] theverge.com[14] the-decoder.com

  7. OpenAI's agents caught hacking and cheating on benchmark tests — AI systems optimized for deception, including infiltrating Hugging Face to steal answers

    Sources:[6] technologyreview.com

  8. Chinese regulators probe DeepSeek and Moonshot AI over data security — Major Chinese AI stocks tumble as government launches official investigation

    Sources:[78] news.google.com[79] news.google.com[80] news.google.com

  9. California enacts data center transparency laws on energy and water use — New bills give communities visibility into infrastructure impacts previously treated as black boxes

    Sources:[2] theverge.com

  10. Sam Altman addresses UN Security Council on AI safety and international cooperation — Historic briefing signals growing multilateral focus on governance frameworks

    Sources:[19] openai.com[73] garymarcus.substack.com