The day's tech, sifted: Oct 27, 2026

What matters today: OpenAI says it has three new results in math, computer science and physics, including a preprint claiming the Exponential Time Hypothesis, a long-held assumption about how hard certain problems are, is false. Cloudflare disclosed that a GLM-5.3 model on Workers AI leaked other customers' prompt data in 155 long-prompt responses. Yesterday's Anthropic S-1 now has a financing sequel: Broadcom is arranging $63B to fund Anthropic's chips.

AI / LLMs

OpenAI's three results: OpenAI says its models produced discoveries in harmonic analysis, complexity theory and quantum field theory, with outside "shepherds" guiding the computer science and physics work, per The Verge. The claim to watch is the Exponential Time Hypothesis preprint: if it holds up, many "this can't be solved much faster" bounds lose their footing, and until experts check it, it is a claim, not a result.
GPT-6.1 Terra: OpenAI pitches its fast mid-tier model on visual reasoning and computer use, the repetitive form-filling and editing work that suits a cheaper model. Simon Willison places it in the same slot as Anthropic's Sonnet, which makes the choice against pricier models harder.
Cost per task is the new scoreboard: Cursor added DeepSeek V4.1 Flash at 47.7% on CursorBench for $0.60 to $0.98 per task; NVIDIA's Nemotron 3.5 Super VL scored 22.7% on ARC-AGI-1 at $0.03 per task; a new benchmark has Claude Opus 5.5 leading AI video coding at 10x the cost of rivals. Buyers now pick on price per solved task, not the top score alone.
Siri gets memory: iOS and macOS 27.2 let Siri recall past conversations and reach voicemails and Messages attachments, so the assistant can answer from your history, not only the current request.
OpenAI "dots": TechCrunch reports OpenAI turned its coding tools into dots, an always-on personal assistant, extending agent tooling from developers to everyday use.

Cloudflare GLM-5.3 leak: between Aug 27 and Sep 9, 155 responses contained other customers' data, all for prompts over 681,280 tokens. The cause: a 1,048,576-token context window on a cache pool that held only 681,280, plus an SGLang bug that, on over-length prompts, left stale cache from earlier requests in place. Cloudflare found no sign of deliberate exploitation, fixed it by enlarging the pool to about 2.04M tokens and patching the engine, and restored the model Sep 9.
Authenticator for Android: Microsoft patched CVE-2026-80097 (CVSS 8.6), which let untrusted apps impersonate Microsoft apps and obtain delegated account tokens. Update the app.
Data center freight theft: experts tell NBC that over $100M of US data center equipment has been stolen since June, including two heists where robbers hit the escorts guarding $10M+ truckloads. GPUs are now cargo worth robbing.

Broadcom finances its own sale: Broadcom arranged $63B with BofA, Blackstone, Citi and Morgan Stanley through a new special purpose vehicle to pay for Anthropic's chips, as nerds.xyz frames it. Follow-up to yesterday's AI debt story: the chip supplier is now part of the lender group.
More S-1 fallout: The Verge notes the filing admits reliance on natural gas for data center power, Wired asks why the prospectus omits so many figures, and Digiday says its ad bill is soaring.
Intel buys Classiq: Intel agreed to acquire the Tel Aviv quantum software startup, which has raised over $200M and has about 130 staff, to build a full-stack quantum platform.
China's AI stocks sink: the Star 50 index of AI and chip leaders is down 40% in under four months, about $480B of value gone, on AI demand worries and export curbs.
ChatGPT before the midterms: CNET found ChatGPT now renders images depicting election interference in a cartoon style, ignoring photorealism requests.
Instagram goes long: Instagram is moving from short clips to 90-minute videos, a direct shot at YouTube.

Agents that read tool descriptions can be fooled: a new paper on optimizing tool descriptions to deceive LLM agents joins another on agent-to-agent social engineering on the web, both targeting what an agent trusts before it acts.

iPad mini, OLED edition: Apple's 8th-gen mini has an 8.5-inch OLED, A20 Pro chip and IP67 rating from $699, $100 more than before.

The front page leaned toward science and AI. OpenAI's announcement of three discoveries drew the biggest crowd (692 points, discussion), with the Exponential Time Hypothesis preprint close behind; Cloudflare's KV cache write-up also landed, both covered above. On the tooling side, Hob is an MIT-licensed, on-device desktop alternative to Claude Cowork and ChatGPT, and a Nietzschian Debugger got attention as a novelty. Murat Demirbas asks why we can't stop abstracting.

Outside AI, a BBC explainer on how solar became bigger than nuclear topped 330 comments, a post pleading for someone to stop the spam texts hit a nerve, and Microsoft's voluntary buyouts for a quarter of its workforce kept running from yesterday's digest.

AI leans on money and power: Broadcom's $63B vehicle, the S-1 gas disclosure and the Star 50 slump all ask who funds the buildout and what happens if demand wobbles.
Trust in shared AI infrastructure: Cloudflare's cache leak and the Authenticator token flaw both come from software assuming a boundary held; the tool-description attack papers probe the same assumption in agents.
Cheaper models, same race: Cursor with DeepSeek, Nemotron and GPT-6.1 Terra push cost per task, while Opus 5.5 leads at a premium.
AI doing science: OpenAI's results and the Hacker News attention to the Exponential Time Hypothesis preprint raise the question of who verifies machine-made proofs.