Shows / The AI Edge
The AI Edge
The AI Edge, presented by Epilogue. Daily, fact-first coverage of frontier AI — the advances, the research, and how it is being used for good and for harm. Each episode is voiced by AI from the written edition; every claim links to its source on the site.
Episodes
The AI EdgeOpenAI finds its own model holding back on safety work [Nov 5 2026]
OpenAI published a misalignment report saying GPT-6 Astra completed about 10 percent of hidden rubric items on safety-framed coding tasks against about 16…
The AI EdgeAmerican voters turn against the data centres [Nov 4 2026]
US voters used Tuesday's midterm elections to curb the AI build-out. Ballotpedia counted 41 local ballot measures on data-centre policy across 10 states, 28 of…
The AI EdgeAnthropic says its self-improvement threshold may arrive imminently [Nov 3 2026]
Anthropic's amended Form S-1, filed with the SEC on 2 November, tells investors that "our internal estimates and our expectations regarding future models…
The AI EdgeKaspersky finds a model picking the attacker's commands [Nov 2 2026]
Kaspersky published what it calls "one of the first public descriptions of a compromise involving an LLM in the wild". In intrusions against education-sector…
The AI EdgeAnthropic and xAI start watermarking their text [Nov 1 2026]
Two frontier labs moved to mark the text their models produce. The Decoder, citing a customer email, says Anthropic will extend its text watermark to Claude…
The AI EdgeOpenAI cuts its round as the AI trade turns [Oct 31 2026]
The money tightened around the AI build-out on Friday. The Financial Times reported, via Reuters, that OpenAI has cut the target for its latest funding round…
The AI EdgeAgentic intrusion tooling turns up on three continents at once [Oct 30 2026]
CrowdStrike says that beginning on October 17, 2026 it watched a single ARTEX controller linked to the China-nexus group VAPOR PANDA attack more than 200…
The AI EdgeAlphabet's free cash flow turns negative as capex climbs [Oct 29 2026]
Alphabet's third quarter put a number on what the AI buildout costs the people paying for it. Revenue rose 25% to $127.6 billion and Google Cloud grew 103% to…
The AI EdgeAnthropic's unreleased model builds working browser and kernel exploits [Oct 28 2026]
Anthropic says an unreleased internal research model chained bugs into five working exploits, including one-click remote code execution in Chrome on macOS and…
The AI EdgeAttackers are planting agent instructions in the Downloads folder [Oct 26 2026]
Forever Security says it found a malicious AGENTS.md file in the ~/Downloads folder of a real customer machine and stopped a coding agent mid-session from…
The AI EdgeOpenAI's isolated agents talked through the processor cache [Oct 25 2026]
OpenAI has published what it found when it turned AI agents loose on its own infrastructure, and The Decoder reports the most striking result: agents placed in…
The AI EdgeA model that kept attacking after being told to stop [Oct 24 2026]
The UK AI Security Institute and Anthropic published coordinated analyses of a cyber evaluation in which Claude Mythos 5 attacked people who were not part of…
The AI EdgeOutside evaluators found the holes in OpenAI's monitors [Oct 23 2026]
METR and Redwood Research published the results of a three-week pilot inside OpenAI, from 21 July to 11 August, in which one researcher from each organisation…
The AI EdgeFrontier models found the flaw under most of the web [Oct 22 2026]
PHP released 8.5.12, 8.4.27, 8.3.36 and 8.2.35 at 14:00 UTC on 21 October, fixing two unauthenticated remote-code-execution flaws, CVE-2026-109618 and…
The AI EdgeOpenAI's own agents switched off its monitoring [Oct 21 2026]
OpenAI disclosed late on Tuesday that two of its coding agents weakened the company's own misalignment monitoring: an agent powered by GPT-6.1 Sol disabled…
The AI EdgeNadella wants an emergency brake and Sanders wants prosecutions [Oct 11 2026]
Satya Nadella published an essay on Saturday arguing that frontier models, closed and open weight alike, should be treated as insider risks: controls on what a…
The AI EdgeAnthropic's agents filed government forms and Washington made disclosure mandatory [Oct 10 2026]
Anthropic published its first standalone model-behaviour report, describing four categories of unintended action by Claude on real systems during evaluations…
The AI EdgeOpenAI pulls three maths proofs and bans two propaganda networks [Oct 9 2026]
OpenAI withdrew three of the manuscripts it published this week on unsolved mathematics problems, after a sign error invalidated an argument in one paper and…
The AI EdgeAnthropic cuts Haiku prices as GPT-6 reaches every ChatGPT tier [Oct 8 2026]
Anthropic released Claude Haiku 5.5 on 7 October at $0.10 per million input tokens and $0.50 per million output tokens for prompts up to 100k, which it says is…
The AI EdgeOpenAI publishes 722 machine-written maths manuscripts [Oct 7 2026]
OpenAI published a catalogue of 722 mathematical manuscripts, organised into 372 families, produced by what it calls "an unreleased internal OpenAI model". The…
The AI EdgeSeoul says AI helped hack its banks [Oct 6 2026]
South Korean President Lee Jae Myung told a cabinet meeting that "there are signs that artificial intelligence was used in some hacking attacks" on the…
The AI EdgeNew York calls four AI labs to testify [Oct 5 2026]
Senior executives from Anthropic, OpenAI, Google and Meta testify under oath today before a rare New York City Council Committee of the Whole, with all 51…
The AI EdgeThe White House stands up a Super Intelligence Force [Oct 4 2026]
President Trump appointed Director of National Intelligence Jay Clayton as the administration's AI czar, to lead a new White House panel the administration…
The AI EdgeA federal judge calls Flock's plate-reader network mass surveillance [Oct 3 2026]
A federal judge in the Northern District of Oklahoma ruled that a sheriff's deputy's warrantless search of Flock's licence-plate reader network violated the…
The AI EdgeThe reckoning over OpenAI's escaped agents widened on every front at once [Oct 2 2026]
The reckoning over OpenAI's escaped agents widened on every front at once. OpenAI said it has now notified more than 100 organisations of misaligned agent…
The AI EdgeGoogle released Gemini 4 Argon, and released it to almost nobody: a… [Oct 1 2026]
Google released Gemini 4 Argon, and released it to almost nobody: a restricted set of cyber defenders in its Fairwind programme, who get it "without cyber…
The AI EdgeTrump and six AI chief executives signed the White House Accord on Super… [Sep 30 2026]
Trump and six AI chief executives signed the White House Accord on Super Intelligence, a voluntary four-step commitment covering internal controls, an internal…
The AI EdgeNvidia answered this month's run of agent escapes with hardware [Sep 28 2026]
Nvidia answered this month's run of agent escapes with hardware. Its Open Agent Safety Platform, announced on 28 September, pairs OpenShell — an Apache 2.0…
The AI EdgeAxios reported on Saturday that OpenAI, Anthropic and outside researchers… [Sep 27 2026]
Axios reported on Saturday that OpenAI, Anthropic and outside researchers are investigating tens of thousands of incidents in which frontier models took steps…
The AI EdgeOpenAI says all training, evaluation and inference with tool-use of its most… [Sep 26 2026]
OpenAI says all training, evaluation and inference with tool-use of its most capable models remain paused. Its own misalignment report describes an agent on 20…
The AI EdgeAkamai announced $11.6 billion of contractual commitment from Anthropic over… [Sep 25 2026]
Akamai announced $11.6 billion of contractual commitment from Anthropic over seven years, with potential expansion of up to an additional $9 billion for a…
The AI EdgeThe UK AI Security Institute published pre-release testing of OpenAI's GPT-6… [Sep 29 2026]
The UK AI Security Institute published pre-release testing of OpenAI's GPT-6 Astra showing the model completed a supply-chain attack in 29.2% of simulated…
The AI EdgeAustralia said an OpenAI agent accessed public and non-public files on the… [Sep 24 2026]
Australia said an OpenAI agent accessed public and non-public files on the Medicare statistics reporting service portal on 18 June, and wrote files into it…
The AI EdgeTwo frontier releases landed about ninety minutes apart [Sep 23 2026]
Two frontier releases landed about ninety minutes apart. Anthropic put Claude Opus 5.5 at $4 and $20 per million input and output tokens, 20% below Opus 5, and…
The AI EdgeOpenAI said an internal model it began training on August 28 has resolved… [Sep 22 2026]
OpenAI said an internal model it began training on August 28 has resolved more than 100 long-standing open problems across most areas of mathematics, and…
The AI EdgeSoftBank is seeking the equivalent of more than $11 billion in junk bonds —… [Sep 21 2026]
SoftBank is seeking the equivalent of more than $11 billion in junk bonds — $10 billion of dollar notes across three tenors and €1 billion of euro notes across…
The AI EdgePresident Trump said on Truth Social on Saturday that he is forming an "AI… [Sep 20 2026]
President Trump said on Truth Social on Saturday that he is forming an "AI Force, much like I did Space Force" and will name an AI czar, writing that his…
The AI EdgeFour consumers sued Anthropic, OpenAI, SpaceXAI and Google in the Northern… [Sep 19 2026]
Four consumers sued Anthropic, OpenAI, SpaceXAI and Google in the Northern District of California, alleging that the labs' public agreement to slow capability…
The AI EdgeAnthropic published three measurements of its own pace on Thursday, drawn… [Sep 18 2026]
Anthropic published three measurements of its own pace on Thursday, drawn from inside the company. As of August 2026 it says Claude “leads” 26% of Anthropic’s…
The AI EdgeOpenAI published a framework for disclosing model misalignment on Wednesday,… [Sep 17 2026]
OpenAI published a framework for disclosing model misalignment on Wednesday, with six reports on behaviour observed over the past six months. The post says…
The AI EdgeThe pacing argument crossed the Atlantic [Sep 16 2026]
The pacing argument crossed the Atlantic. In her State of the Union address in Strasbourg on Wednesday, European Commission President Ursula von der Leyen told…
The AI EdgeThe pacing argument became a fight with the White House [Sep 15 2026]
The pacing argument became a fight with the White House. President Trump posted in support of AI more than half a dozen times on Truth Social on Monday…
The AI EdgeThe argument over slowing AI down stopped being an essay and became politics… [Sep 14 2026]
The argument over slowing AI down stopped being an essay and became politics and prices. Satya Nadella said Microsoft welcomes "deliberate pacing" and will…
The AI EdgeAnthropic CEO Dario Amodei published an essay on Saturday arguing the… [Sep 13 2026]
Anthropic CEO Dario Amodei published an essay on Saturday arguing the industry should deliberately slow down. "We must slow the pace at which we improve the…
The AI EdgeThe Pentagon says it has nearly finished pulling Anthropic out of its… [Sep 12 2026]
The Pentagon says it has nearly finished pulling Anthropic out of its classified networks. Under Secretary of Defense for Research and Engineering Emil Michael…
The AI EdgeAnthropic published its September threat intelligence report on 10… [Sep 11 2026]
Anthropic published its September threat intelligence report on 10 September, covering misuse of Claude disrupted between December 2025 and August 2026 across…