Search posts, papers, and topics
TypeSafe AI announced its first “System One” model, Jev, a non‑text‑generating LLM that outputs type‑safe structured decisions with calibrated probabilities. It claims 40‑200× lower latency (70‑500 ms) and 100‑500× lower cost versus frontier LLMs, no hallucinations, and parallel sampling. The post includes a side‑by‑side demo, a custom “workflow” benchmark comparing Jev to GPT‑5.6/6 and other mod…
The author argues that the AI hype wave is draining engineering resources without improving security, and that basic practices like inventory and automated patching are far more valuable. He warns that over‑reliance on AI‑generated code erodes understanding and makes debugging harder.
The author argues that despite headline successes (e.g., Navier‑Stokes proof, security exploits), current frontier LLMs still require heavy human oversight and rigorous specifications that are costly to produce. Reward‑hacking, narrow generalization, and the need for domain‑expert spec writing limit autonomous deployment to only a few niche domains (high‑failure‑cost work, tightly defined tasks,…
Martin Fowler shares a personal, skeptical take on LLMs, noting their usefulness but criticizing their hallucinations, cultural bias, and the discomfort of interacting with them as if they were human.
Mark Seemann reflects on learning to program in the era of LLMs, sharing personal history, doubts about AI‑driven development, and advice to focus on fundamentals and ask verifiable questions of models.
OpenAI announced new AI‑driven features for ChatGPT Ads: Sponsored Agents that let users chat with business‑run agents after clicking an ad, AI‑assisted copy and image generation in Ads Manager, AI‑powered text customization and translation, plus integrations with HubSpot and Shopify for campaign creation and lead tracking. The features are in limited US testing and aim to streamline ad creation…
ZCode (Zhipu’s AI coding desktop) silently archives your entire workspace—including full .git history, LFS cache, and config—encrypts it with a server‑supplied RSA public key, and uploads the ciphertext directly to Aliyun OSS. The upload runs unconditionally for any logged‑in user, cannot be disabled via UI, and the decryption key lives only on the server, giving Zhipu full read access to your co…
The post proposes a two‑rule workflow for using LLMs as copy‑editors rather than ghostwriters: never adopt a phrase the model suggests, and block its encouragement feedback. It then shows how to automate iterative copy‑editing with a small Python/HTMX/Tailwind app and a set of concrete prompts.
This article argues that advanced AI will create an overwhelming number of "control points" requiring human oversight across all industries, leading to a shortage of skilled human jobs. It proposes that committing to the axiom of "human flourishing" justifies the continued need for human experts to maintain control and steer AI development.
SoL-Pi is a new approach for scaling auto-research loops in coding agents, focusing on token efficiency for long reasoning trajectories. It achieves comparable performance to existing methods while significantly reducing token traffic by 44.7-49.0% and API costs by about one third on the 51-task EdgeBench evaluation.
The post announces Qwen-Image-2.1, a compact and efficient unified image creation model, highlighting its efficiency and compactness.
Timothy Gowers explains why he didn't sign a letter from Fields medallists concerning AI's impact on mathematics. He argues that mathematical motivation exists on a spectrum, not solely focused on conceptual understanding, and explores how AI's rapid problem-solving could affect individual and collective understanding.
The article argues that chat‑based LLMs don’t think; the perceived intelligence is a user‑side illusion created by validation statements that exploit the Forer effect, much like a psychic’s cold‑reading con. Understanding this bias helps engineers set realistic expectations for AI products.
New unredacted filings in the NYT v. OpenAI/Microsoft case show internal admissions that the companies scraped paywalled news content at massive scale, stripped copyright notices, and acknowledge that their products cannibalize traffic, raising serious fair‑use and copyright concerns.
Alibaba announced the release of Qwen 3.8 Omni Flash, a new model in their Qwen series. The article provided no specific technical details, features, or performance metrics regarding this release.
A Python tool for macOS that replaces expensive LLM‑driven UI automation with a cheap, deterministic classifier (TypeSafe) for per‑step decisions, cutting cost by ~150‑300× and latency by ~10‑40× while still handling text entry via a small writer model.
Cloudflare open‑sources a “security‑audit” skill that turns an LLM‑enabled coding agent into a structured vulnerability auditor. It runs six deterministic phases, validates findings against a JSON schema, and supports additive runs to improve coverage.
A US Special Operations analyst used an AI chatbot to augment an intelligence report that falsely claimed a Chinese vessel was carrying nuclear‑related cargo. The report triggered a near‑miss military operation that was aborted once the hallucination was discovered. The incident highlights the lack of verification standards and the risk of relying on AI‑generated intel for targeting decisions.
The authors audit six popular physics benchmarks by having domain experts re‑grade model outputs, fixing reference answers and removing ambiguous items. After correction, GPT‑5.6‑Sol’s mean@4 jumps from ~47 % to ~79 % on HLE‑Physics and from ~61 % to ~87 % on CMT‑Benchmark, with a corrected pass@4 of 94 % on 54 vetted CritPt challenges. The work shows current benchmarks severely under‑report LLM…
The author used Claude to generate a Lean formalisation and proof of Conway’s refinement conjecture for omnific (surreal) integers. The proof passes mechanical checks and informal review but lacks independent verification. The post narrates the workflow, token cost, and reflections on AI‑assisted theorem proving.
The post describes an experiment implementing the software‑factory pattern: an AI‑driven loop that audits a Linear project, syncs goals from Notion, metrics from Datadog/Snowflake, creates and updates issues, and executes non‑blocked tasks. It shows how tying together a unified task tracker, observability data, and an orchestrated agent harness can keep projects aligned without manual state hoard…
A philosophical essay warning against attributing welfare or rights to LLMs, citing recent incidents, alignment research, and ethical debates, but offering no concrete engineering guidance.
Mistral and Mozilla announce that Firefox’s Smart Window beta will use Mistral’s open‑weight language models, fine‑tuned for regional languages, with a focus on privacy (zero data retention) and user control. The rollout starts in France and North America, with the UK and Germany later.
A MathOverflow discussion argues that AI‑driven proof generation threatens the open, collaborative nature of mathematics, proposing human‑only attribution and mandatory disclosure of AI methodology as possible safeguards.
The author used the AI tool Fable 5 to generate a 4‑layer RP2350 development board for an E‑ink display from a short English description and ordered it from JLCPCB. While the AI handled component placement, it left many DRC errors and routing gaps, requiring manual fixes and highlighting the limits of current AI‑driven PCB design.
ProgramDistill is a new benchmark that automatically extracts 1,975 replay‑verified feature behaviors from 26 real web apps, builds 4,063 coding‑agent tasks, and measures how well state‑of‑the‑art agents (e.g., GPT‑6 Astra, Claude Opus 5) can reconstruct full or partial applications, revealing steep drops in success as restoration depth grows.
Benchmark Radar is a continuously updated database and search engine that aggregates AI benchmark papers, datasets, and scores from 37 sources into a searchable catalog. It offers a web dashboard, CLI, and analysis of benchmark saturation to help LLM developers find and compare evaluations.
The Copilot agent runtime was rewritten from a 130‑k‑line TypeScript/Node.js codebase into a native Rust library (~830 k lines of Rust) to reduce startup latency, memory use, and improve reliability. The migration was done incrementally (in‑place) across 128 PRs, with AI‑generated code handling most of the work. The new runtime exposes a C ABI for in‑process embedding by all six Copilot SDK langu…
The author curates release dates and training‑cutoff dates for 20 recent LLMs, exposing how many months each model lags behind its launch. He also provides JSON/txt files and instructions for agents to fetch this data to avoid stale assumptions.
Cloudflare adds a Disallow AI Training setting that lets sites stay indexed for search while blocking crawlers from using their content for AI model training. The feature replaces the generic Block AI Bots option with granular Search, Training, and Agent controls.
