Sunday, 27 September 2026
This dated roundup collects the most interesting AI and technology developments found for Sunday, 27 September 2026.
Policy & Ethics
Anthropic Investor Lonsdale Warns AI Safety Push Could Entrench Leading Labs
He accepts that AI safety risks are real but argues that leading developers’ influence over regulation could strengthen their market position.
Read moreKākāpō Party
Tool: Kākāpō Party I presented a closing keynote for the WeAreDevelopers World Congress North America yesterday. As a STAR moment I decided to weave in references to the record breaking kākāpō breeding season we had in 2026. For my closing slide I wanted to celebrate, and I had seen some buzz around how good Claude Opus 5.5 was at creating pixel art animations. So I rounded up three Kakapo photos from Google image search and dropped them into Claude with this prompt: Here are some photos of kakapo parrots just to remind you what they look like I need you to make an animation in animated...
Read moreAustralian Senate Inquiry Calls OpenAI and Anthropic Chiefs to Answer AI Safety Questions
The invitations are voluntary for now. Lawmakers want public answers after an OpenAI test agent gained unauthorized access to a government statistics portal.
Read moreResearch & Products
Anthropic Releases Lower-Priced Claude Opus 5.5 With a Fable-Level Performance Claim
The token rates are clear. Whether the model matches the more expensive Fable 5.1 on a customer’s work is a separate question.
Read moreSupersonic Labs Releases CPU-Friendly AI Router With a Weakness on Crowded Choices
Julia 1 can choose among supplied answers without generating text. Supersonic’s tests suggest long lists of similar options remain a problem.
Read moreGen Launches AI Privacy Checker That Starts With an Email Address
The beta explains what listed services’ policies allow, including AI training and data retention. Gen’s launch materials do not say how it handles the identifier users enter.
Read moreTags: hardware
Google DeepMind Engineer Quits, Saying His Chip Work Could Speed Up AI
Robert O’Callahan says his tools could make AI cheaper and faster. He sees the risks as uncertain, but serious enough to change his work.
Read moreTags: ai_agents
Meta Lets Muse Users Request Early Access to Features Previewed at Connect
A message to the AI agent puts users forward for a possible trial, but it does not unlock the video, desktop or glasses features Meta has previewed.
Read moreMeta’s smart glasses take center stage at Connect, with audio-only models and AI agent Muse
Meta spotlighted smart glasses at Connect, unveiling audio-only models, hearing support glasses and tighter Muse AI integration. The post Meta’s smart glasses take center stage at Connect, with audio-only models and AI agent Muse appeared first on Superintelligence News - Artificial Intelligence News .
Read moreTags: announcement
Guardian Finds Google’s India AI Project Cleared for 2.51GW, Versus Announced 1GW
Three petitions challenge the project’s environmental approvals, while residents and the state dispute whether land was taken against their will.
Read moreTags: ai_safety
The AI safety debate is confusing. Here's our guide to the different factions
The battle over AI's promise and perils is not a clash between two opposing sides, but a debate among a range of competing voices and ideologies. Here is a guide to who's who in that debate.
Read more