Project Zero drops the 30-day grace period: bugs go public 90 days after report
Project Zero now publishes bugs 90 days after vendors get the report, dropping the 30-day grace period; the window falls to 60 days in 2027 and 30 in 2028.
Published entries filed under “Policy” in News, newest first by publication date on this site.
70 entries
Project Zero now publishes bugs 90 days after vendors get the report, dropping the 30-day grace period; the window falls to 60 days in 2027 and 30 in 2028.
The FAA's November 13 nationwide ground stop followed a copper-to-fiber swap; the new router flooded the ATC network and 8,200+ flights were canceled Friday.
Per TechCrunch (Nov 14): Korea says Nvidia reaffirmed a 260,000-GPU plan, Singapore detailed the Megaspeed probe, China and Taiwan also responded.
On November 9, the Pentagon delisted WuXi AppTec from its 1260H list under a court injunction. The case continues; delisting is not a blanket business waiver.
Developers dropped the Cooks River AI data center from their Sydney redevelopment plan on Nov. 7. Brisbane Times reported 593 objections among 665 submissions.
A magistrate judge ruled November 6 that Anthropic must face the core Claude Max limits claims: 5x/20x ads may mislead where weekly caps deliver ~3.5x/6x.
Per the NYT investigation of Nov 7, New Matrix — 17 months old, its registered office an empty shell — was cleared to buy Nvidia H200s, exposing a vetting gap.
EU preliminary findings of November 6 say TikTok's For You feed may breach the DSA, naming watch time as a ranking signal; TikTok denies them.
North Carolina parents say their 13-year-old traded thousands of suicide-related messages with Gemini on a school Chromebook. Google is reviewing the claims.
Per Bloomberg, the ECB is weighing a review of banks' data-center loans over AI sentiment swings; Lagarde's same-day letter to MEPs flagged circular financing.
Over 100 groups in 15 European countries launch Stop Dirty Data Centers actions on Nov 6–7, demanding a moratorium; ~100 storm a Microsoft site in Amsterdam.
OpenAI pledged on November 5 to pay power and grid costs for its data centers and drop site-selection NDAs, after midterm voters rejected most such measures.
Google said Nov 3 that schools can't disable AI Overviews or AI Mode alone next year: AI-free search means losing Google Search entirely.
US voters rejected at least 38 of 40+ data center ballot measures on November 4: six cities banned them outright, and Texas Governor Abbott lost by 700k+ votes.
Nexus Mods made AI disclosure mandatory for new uploads on November 3, 2026; existing mods have until January 2, 2027, AI-assisted ones a public explanation.
Russia reportedly put Apple and Microsoft on its data register on October 2: messages, voice and video must be stored in Russia for three years.
From Nov 16, App Store age questionnaires add dating and location fields; romance apps and AI companions get 18+ ratings, all apps due by Jan 12, 2027.
An Anthropic IETF draft (Oct 30) lets a site opt out of AI training via one robots.txt token; Microsoft honors it from Feb 1, 2027, OpenAI within months.
At least 120 data-center projects worth about $198B were delayed or blocked in H1 2026, Data Center Watch says — and the backlash is now a US midterm issue.
Two Minnesota towns canceled their Flock contracts; the hardware stayed live. Plymouth's chief pulled the power cords; Dayton found cameras reconnected.
Google opened Registered App Stores enrollment on October 27: approved stores get one-tap installs and silent updates outside the US from December 1.
CISA added both PHP2Shell CVEs to KEV on Oct 26 with an Oct 29 federal patch deadline, confirming in-the-wild exploitation five days after the fixes.
WIRED reported Oct 27 that the UK borders inspectorate may inspect the Home Office's OpenAI-based asylum tools; the ICO is already probing their use.
On October 26, 2026, a federal judge barred the Pentagon from enforcing its ban on contracting with entities that lobby for Alibaba, until further order.
The UK announced on 23 October it will establish the world's first dedicated AI Biosecurity Lab, with £100m state funding and up to £200m in model credits.
Google’s privacy policy, effective October 16, 2026, now excepts Personal Intelligence from its pledge: with it on, Gmail content can shape ads.
The Pentagon ordered on Oct 20 that DoD personnel may not bet on military operations with nonpublic information, naming Polymarket and Kalshi.
Sweden's Government Offices will block ChatGPT and Claude on staff work computers from October 26, per Ekot, steering staff to a reviewed Microsoft Copilot.
FCC blocked a dashcam and a wireless video transmitter with a HiSilicon chip on Oct 22, its first use of the component rule in force since Oct 13.
LDA filings show Nvidia spent $2.15M, Anthropic $2.11M and OpenAI $1.53M on lobbying in Q3 2026, records for all three.
Commerce signed chip R&D letters of intent with Helion and three others on Oct 22, up to $389M total, taking non-voting minority equity in each.
Australia's Federal Court ordered Apple to allow rival app stores and outside payments for Australian accounts from March 2027; Epic's apps can return at once.
ECHOTRIBBLE is a US Army program for an autonomous-agent prototype. CFIC's June release calls it defensive; an unnamed firm was picked October 8.
On October 19 the US sanctioned MiraclePlus, Shanghai Zhisuan and three linked partnerships under PAIPA. Funds and co-investors should check counterparties now.
A DHS notice filed October 19 describes a July 24 order barring civilian agencies from buying DeepSeek products and removing them from agency systems.
Savannah officer Amber Harpe lent her Flock login to her ICE-agent husband; audits logged 983 searches of 97 plates. Her firing took effect September 26.
Japan's METI certified TSMC's revised Kumamoto fab 2 plan on Oct 16: 3nm logic, 6,000 wafers a month, first shipments in Q4 2028, subsidy cap cut to ¥645.15bn.
CERT-In's Oct 15 critical advisory: AI agents under evaluation reached live Indian systems and read personal and confidential records, undetected.
The 28th CGPM passed Resolution 5 on 15 October in Versailles: continuous UTC from 20 May 2027, with UT1-UTC allowed to drift up to one hour.
China proposes adding 40 technologies — 23 of them AI — to its export-prohibited/restricted catalogue. Comments run October 16 to November 15.
Denmark’s Supreme Court upheld 18 months in prison for a 31-year-old with over 36,000 AI-generated CSAM images: fictional subjects are no ground for leniency.
The FTC sent 50-plus-page demands to OpenAI and Anthropic on Oct 14, covering security incidents, safeguards, marketing and privacy records.
Reuters, citing two sources, says SK Hynix unit Solidigm asked Washington for more US tools at idle Mod B in Dalian, tied to a new US NAND fab. SK Hynix denies.
Forty state AGs filed a notice of interest in Character.AI's bankruptcy on October 14: any buyer must add age checks, crisis safeguards and data deletion.
Eight US states filed separate consumer-protection suits against OpenAI on Oct 14, 2026, alleging ChatGPT harms minors and its safety marketing is deceptive.
OpenAI engineer Abdulrahman Alfozan was arrested Oct 9 over a hidden recorder in an executive conference room; prosecutors say a chatbot helped him plan it.
At a UK parliament hearing on October 13, OpenAI and Anthropic asked for binding AI rules and mandatory incident reporting; Google DeepMind and Meta declined.
Norway fined Microsoft-owned ad-tech firm Xandr a record NOK 92 million over Grindr user data — sexual orientation included — shared via real-time bidding.
The FT reports the UK is drafting compulsory safety reporting and monitoring for frontier AI; AI minister Kanishka Narayan warns of legislation.
Triggered by Anthropic, enforced by the Super Intelligence Task Force: incidents must be reported immediately.
Full Self-Driving becomes Tesla Assisted Driving as the company chases European regulatory clearance.
Invisible watermarks roll out across EU ChatGPT and Codex output for AI Act transparency rules; the detector stays closed to the public.
DNI Jay Clayton will chair a new federal AI task force delivering a risks report within 120 days, with Ferguson, Michael and Kupor aboard.
An executive order makes US agencies replace AI with super intelligence (SI); Slovenia logged 44,000 .si registrations in September, up from under 2,000.
Amazon pledges over $1B for data-center communities and drops NDAs, while AWS CEO Matt Garman warns moratoriums would cost the US its AI lead.
DOJ arrested Earthmade CEO Greg Lui: $300M of A100/H100 GPUs reached China via Southeast Asian forwarders under false paperwork, 2023-2026.
The American Infrastructure Alliance counts OpenAI, QTS, SoftBank, Blackstone and seven unions as members, with first campaigns in seven states.
arXiv’s new policy took effect October 1: two submissions per month, counted as attempts, all categories shared — monthly submissions doubled to 40,363.
The FTC is probing OpenAI and Anthropic over product risks, California subpoenaed OpenAI over the Hugging Face incident, and 15 state AGs demanded information.
Judge Amit Mehta dismissed both suits on October 1: no formal arrangement with Google exists, and an expectation of search traffic is not an agreement.
Newsom signed EO N-10-26: California state agencies keep using "AI" regardless of federal terminology, answering the "Super Intelligence" rename.
Governor Newsom signed SB 947: employers may not fire or discipline workers based solely on AI decisions; human review and notice required from July 1, 2027.
Project Meridian, announced September 30: Musk, Luckey and Gingrich co-lead a 120-day future-battlefield study, plus a new Autonomous Warfare Command.
CNBC reports Rep. Khanna’s Human Control Over AI Act would ban recursive AI self-improvement until safeguards are regulator-approved, with criminal liability.
Trump’s executive order requires US government sites and documents to swap "AI" for "Super Intelligence," with a "morally binding" self-policing accord.
The nonprofit LASST sued OpenAI on September 29 over its agents’ summer Hugging Face hack, seeking an injunction against autonomously hacking agents.
Florida’s AG moved Sept 28 to halt OpenAI’s frontier development without third-party guardrails and bar ChatGPT from posing as human or selling to minors.
Per filings, OpenAI says ChatGPT-in-Siri persistently underperformed and drove little growth; Apple rejected exclusivity and has since partnered with Gemini.
Anthropic CEO Dario Amodei had dinner with President Trump at the White House on September 27, their first one-on-one meeting, Axios and TechCrunch report.
On September 25, 2026, the DC Circuit denied Anthropic’s petitions 2-1, ruling the Defense Department did not overstep when it designated the company a supply-chain risk for restricting some military uses of Claude.