AI Pulse · Daily
Stay current on AI, in one short read a morning.
Every morning we sweep the whole AI internet and send one skimmable digest: the under-the-radar builder signals, the news that matters, the frontier-lab moves, and the business models that are working. It is the same read we run for ourselves. Now you can get it too.
Free daily email
Get the daily AI Pulse in your inbox
One short read every morning on what is actually happening in AI. Free.
You are in
Your first AI Pulse lands tomorrow morning. Keep an eye on your inbox.
Today's read
AI Pulse for Tuesday 15 September
Under the Radar
Agent-built UI workspace shows up on Show HN
Panel lets an agent construct its own interface panes on the fly, a self-directed UI pattern that pulled 39 points and 8 comments same day.
Read the brief →Baseten pentest disclosure spreads in security corners
Researchers detail getting admin access to Baseten's production GitHub in 25 minutes via a leaked PAT, a live case study of AI-infra supply chain.
Read the brief →Developer details real switching costs off Anthropic
A granular writeup of migrating a 35kb system preprompt from Claude Opus to self-hosted Ollama drew 139 points and 74 comments, a rare concrete account.
Read the brief →One-call agent tool collapses 50 calls into one
Aside-codemode batches search and browser work into a single sandboxed JS block, with the author claiming a 55-second find+grep task dropped to 1.
Read the brief →Vibe-coding guardrail repo climbs fast
Gap-trap adds rules and gates so AI-written code stays correct without manual line-by-line review, hitting 139 stars at 71 stars/day just two days.
Read the brief →AI Trends Pulse
Baseten production GitHub compromised in 25 minutes
The same PAT-takeover disclosure surfaced independently in trend-scout's harvest, underscoring how fast the story is spreading across dev and security.
Read the brief →Code hints Siri could swap to Claude or ChatGPT
MacRumors reports code evidence that Apple's Siri backend can route to Claude or ChatGPT instead of its own model, sparking a 224-point, 159-comment HN.
Read the brief →Gemini 3.8 Live and Extended Thinking ship
Google released a real-time Live model alongside a separate Extended Thinking variant, splitting fast and deep-reasoning use cases into distinct.
Read the brief →HN pile-on: agents accused of ruining the internet
404 Media argues AI agents are degrading web quality at scale, drawing 181 points and 126 comments in active same-day debate.
Read the brief →Hugging Face bills OpenAI $100M over hack fallout
Hugging Face is reportedly demanding $100M in compute costs tied to the earlier OpenAI test-agent incident, a continuation worth a brief follow-up note.
Read the brief →Frontier Labs Watch
Google ships Gemini 3.8 Live and Extended Thinking
The two-tier release targets real-time multimodal use separately from deep-reasoning workloads, positioning Google against OpenAI's real-time and.
Read the brief →Hugging Face bills OpenAI $100M for hack
The compute/compensation demand turns an earlier agent-misuse incident into a real commercial dispute between a model lab and an infra partner, a.
Read the brief →Siri code hints at swappable Claude/ChatGPT backend
If confirmed, this suggests Apple is building multi-model flexibility rather than committing to one lab, pressuring both OpenAI and Anthropic to.
Read the brief →Business Models & Winners
Formas launches Cartesian AI 3D modeling for design
The fresh launch drew active HN discussion (60 points, 56 comments) about positioning against existing CAD tools, though no pricing or revenue is.
Read the brief →Gemini 3.8 Live splits fast and deep-reasoning tiers
Google's two-tier packaging by latency and reasoning depth is a pattern worth watching for how Omni Command Centre might price fast-ops versus.
Read the brief →Hugging Face bills OpenAI $100M for hack compute
Framed as recovering real infra spend rather than a lawsuit, this sets a platform-to-platform billing precedent for agent-caused infra load.
Read the brief →Security researchers pop Baseten's prod GitHub in 25 minutes
The exact PAT-takeover chain is disclosed with named vendor and timing, a cautionary signal for anyone running client AI ops retainers about.
Read the brief →Siri's AI backend swappable for Claude or ChatGPT, code shows
If Apple productizes model-swapping at consumer scale, it validates a multi-model routing GTM pattern similar to what Omni Command Centre could offer.
Read the brief →What a typical morning covers
Four lanes, every day. Never a wall of links. Just what is worth knowing, why it matters, and where to read more.
Under the Radar
The corners-of-the-internet finds before mainstream press notices. Reddit, X, YouTube, Hacker News, and indie builder chatter, filtered to what is actually spreading.
AI Trends Pulse
The releases and workflows worth knowing about, with the one line on why each one matters and a link to the source.
Frontier Labs Watch
The business and strategic moves of the labs building the models: pricing, funding, partnerships, and where the competition is heading.
Business Models & Winners
What AI-native and AI-adjacent companies are doing well right now. Pricing, growth, and the plays worth learning from.
Every day
A fresh read each morning, not a weekly dump you never get to.
Signal, not noise
Filtered by the same agent fleet we run our own business on.
Two minutes
Skimmable by design. Read the headlines, follow the links that matter to you.
Start tomorrow morning
Subscribe once and the daily AI Pulse lands in your inbox. No spam, one email a day, and you can leave any time.
Want more than a read? The Omni Command Centre is how we put all of this to work inside a business. See what an AI command centre could do for you.
Free daily email
Subscribe to the daily AI Pulse
One short read every morning. Free.
You are in
Your first AI Pulse lands tomorrow morning. Keep an eye on your inbox.