Tootfinder

You don't want to miss today's Metacurity for the rundown of critical infosec developments you might have missed over the weekend, including
--AI's new battleground: Cost, efficiency, and control,
--AI vendors pivot from capability to cost,
--OpenAI eases GPT-5.6 usage limits,
--Enterprises scrutinize soaring AI bills,
--Chinese AI models gain enterprise traction,
--Anthropic data reveals how AI gets used,
--AI agents tackle business ops w…

AI's new battleground: Cost, efficiency, and control
AI vendors pivot from capability to cost, OpenAI eases GPT-5.6 usage limits, Enterprises scrutinize soaring AI bills, Chinese AI models gain enterprise traction, Anthropic data reveals how AI gets used, AI agents tackle business ops workloads, Open-source AI faces mounting policy pressure, much more

Giants take a step away from mediocrity in Harbaugh's debut win over Cowboys https://www.usatoday.com/story/sports/nfl/giants/2026/09/13/john-harbaugh-cowboys-giants-score-results/91753606007/

Every Saturday, Metacurity rounds up the best infosec-related long reads that deserve more attention than the daily news cycle allows.
This week's collection explores what happens when trust becomes the attack surface, including
• Inside the TfL hack
• The cyber war game nobody wants to play
• AI's toughest test case
• When anti-piracy breaks the internet
• The protocol that changed the internet
• Why prompt injection won't go away

Trust under attack: Best infosec long reads 7/11/26
Inside the TfL hack, The cyber war game nobody wants to play, AI's toughest test case, When anti-piracy breaks the internet, The protocol that changed the internet, Why prompt injection won't go away

RE: https://infosec.exchange/@metacurity/116895832104900222
To be clear, this is a chart made by HackRead and not Darktrace.

The stars must be misaligned because it has been an epic journey to get today's Metacurity out the door, so before you leave for the weekend, don't miss the infosec developments you should know, including
--OpenAI launches GPT-5.6 as AI vendors compete for enterprise users,
--Federal investigators say DOGE records were deleted,
--Europe revives voluntary message scanning,
--Britain plans AI-powered cyber defense system,
--Chinese and Indian hackers targe…

OpenAI launches GPT-5.6 as AI vendors compete for enterprise users
Federal investigators say DOGE records were deleted, Europe revives voluntary message scanning, Britain plans AI-powered cyber defense system, Chinese and Indian hackers target Pakistani police, AI gateway compromise leads to cryptomining, Ransomware negotiator sentenced to 70 mos., much more

AI is giving elite security orgs tools that compress security work from months to minutes. But will these powerful benefits make their way to less-resourced cyber teams?
Check out my latest CSO piece, which examines whether AI can narrow cybersecurity's class divide.
Many thanks to AWS's Steve Schmidt, Blumira's Matt Warner, Google Cloud's Anton Chuvakin, 1Password's Wendy Nather, Kaseya's Dave Baggett, and Ballistic Ventures' Phil Venables for their i…

Can AI narrow cybersecurity’s class divide?
AI is giving elite security orgs new ways to automate red teaming, detection engineering, and vulnerability discovery. But experts debate whether the technology will deepen cyber’s long-standing resource divide — or finally help close it.

EU sanctions Russian GRU military hackers over cyberattacks
https://www.bleepingcomputer.com/news/security/eu-and-uk-hit-russia-with-first-joint-cyber-sanctions-package/

EU sanctions Russian GRU military hackers over cyberattacks
The European Union and the United Kingdom jointly sanctioned dozens of Russian individuals and entities and accused Russia of coordinating a network of hacking groups responsible for attacks across Europe.

The hacker group Black Mirror and the anonymous project VChK-OGPU claimed to have leaked correspondence of Russian propagandist Ksenia Sobchak, exposing how Kremlin censorship works.
h…

Hackers published private messages they attribute to journalist and media executive Ksenia Sobchak, pointing to supposed ‘agreements’ between her media empire and Russian officials — Meduza
On July 8, journalist and media executive Ksenia Sobchak announced that hackers had broken into her email account and briefly gained access to her Telegram channels. Around the same time, screenshots showing fragments of her private correspondence appeared in the channels. Sobchak called them fakes and deleted them. The next day, however, the hacker group Black Mirror and the anonymous project VChK-OGPU published even more screenshots of Sobchak’s messages, along with some of her voice …

South Korea's National Police Agency National Office of Investigation confirmed a large-scale leak of personal access tokens for accessing Github
https://www.chosun.com/english/national-en/2026/07/14/QARGFMJW2VEKVOF7ZUO7TGSCII/

Exclusive: South Korean Police Detect GitHub Token Leak
Exclusive: South Korean Police Detect GitHub Token Leak Over 500 accounts affected; authorities notify Microsoft, Interpol

The FBI and a crew of international partners warn that Russia's FSB continues to exploit poorly configured and vulnerable networking devices worldwide opportunistically
compromising multiple critical infrastructure sector networks.
https://www.ic3.gov/CSA/2026/260713.pdf