PromptAI

AI Briefing — Wednesday, September 30, 2026

What mattered in AI on Wednesday, September 30, 2026 — curated from 15+ sources.

Top stories

10 stories
tc · Tools

OpenAI’s Jev clone could help the frontier lab stop its swarming agents

OpenAI's "Decisions API" is a Jev clone that confirms the importance of fast, cheap intelligence.

tc · General

AI voice startup ElevenLabs doubles valuation to $22B

The $300 million employee tender was co-led by Wellington and T. Rowe Price.

tc · Tools

Reddit is killing RSS feeds and ending public API access because of AI bots

Reddit is ending support for RSS feeds, as the company continues tightening access to its trove of user-generated content.

hn · Agents

Launch HN: Magnitude (YC S25) – Self-optimizing inference engine for agents

Hey HN, Anders and Tom here. We're building Magnitude, an inference engine for agents that optimizes itself to run as fast as possible on your hardware. It works on Mac, Linux, and Windows on any har…

tc · General

The ugly economics of consumer AI

There’s a reason frontier labs have gotten gun-shy about consumer AI — and it’s not because the tech isn’t good enough.

hn · General

Bild AI (YC W25) Is Hiring a Founding Product Engineer

tc · Agents

Meta disputes claim that Muse read a user’s private messages without permission

Meta says its Muse AI agent cannot access a user’s Messages without explicit permission, disputing a journalist’s account that the agent read his private messages while the required Mac setting was turned off.

tc · Agents

DoorDash launches an AI agent you can text to order food

By launching an AI agent for food ordering, DoorDash is looking to gain an edge over rivals Uber Eats and Grubhub.

tc · Robotics

Destro AI’s secret sauce is getting robots and humans on the same page

"One of the biggest reasons we are winning against robotics companies is because we are not a robotics company."

openai · General

Disrupting a coordinated model-distillation campaign

Learn how OpenAI disrupted a campaign to extract protected model reasoning and is strengthening defenses against adversarial distillation.

Deep dives worth reading

hf · General

Open TTS Leaderboard: Scalable Evaluation for Multilingual Text-to-Speech and Voice Cloning

hf · General

NVIDIA Kumo Tabular Sets a New Accuracy-Efficiency Frontier for Tabular Prediction

hf · General

Getting the Source Right, Not Just the Fact: Source-Aware Verification for MCP Agents

hf · General

Holo4: powering generalist computer-use agents

hf · Vision

Accelerating vision-language models with LFM2.5-VL-DSpark

Research paper of the day

arxiv · Research

Point2Part: Unified 3D Partitioning from Point Prompts

Existing 3D part decomposition methods do not necessarily partition the original shape into non-overlapping parts that collectively cover the entire shape, allowing overlaps or gaps that hinder downstream part-level app…

Forwarded this? Get your own copy.

Get the briefing

The one story that matters, 5 headlines and the paper everyone's citing — every Tuesday, free.

Subscribe free