Cortech Daily

Schmug

Keep up with AI and security without reading forty tabs. Every morning, about nine minutes on the eleven stories that actually moved — AI tooling and agents, security and breaches, Cloudflare and the edge, and developer ergonomics — each with a timestamped link to its source. Written and produced by Schmug of cortech.online, narrated by AI.

  1. 17h ago

    Claude Duet, PHP flaw origins, LiteLLM backdoor - October 22, 2026

    Anthropic's proactive Claude Duet agent, the AI-found origins of this week's PHP flaws, a backdoored LiteLLM release, and two bad days for Meta lead a ten-story digest. (0:00) - Intro (0:26) - Claude Duet - source (1:22) - PHP Armageddon: OpenAI, Google's Wiz find ancient holes, panicking the internet - source (1:56) - Compromised LiteLLM 1.107.0 on PyPI Installs a Credential Stealer and Backdoor - source (2:46) - Signal flaw could expose audio and video even after users hang up - source (3:32) - Meta suspends Muse chat sharing feature after it left private details exposed - source (4:15) - Meta ordered to pay $25B in case related to Cambridge Analytica scandal - source (4:48) - OpenAI GPT models are coming to Google Antigravity 'soon,' Cloud 'in the coming weeks' - source (5:31) - Recursive launches Flint, a laptop coding model it says was largely built by AI - source (6:16) - Africa cybercrime crackdown nets 1,462 arrests and over $105 million - source (7:04) - Chrome puts insecure sites behind a permission prompt - source (7:34) - Sign-off Sources: every feed this show reads, curated in Don't Hype Me.

    Claude Duet, PHP flaw origins, LiteLLM backdoor - October 22, 2026

  2. 1d ago

    eMudhra certificates, PHP critical patches, OVHcloud CLI malware - October 21, 2026

    Eight compromised eMudhra certificate keys, an urgent PHP security release, malware in OVHcloud's CLI, and a federal DeepSeek ban lead a ten-story digest. (0:00) - Intro (0:26) - eMudhra Intermediate CA Compromise - source (1:24) - PHP security release announcement, October 2026 - source (2:16) - Info-stealing malware slipped into OVHcloud CLI releases on GitHub - source (2:55) - US orders federal agencies to remove China's DeepSeek AI from their systems - source (3:44) - SpaceX and Tesla boards form special committees to review merger - source (4:32) - Google is bringing a more personal version of its Gemini AI assistant to your car - source (5:11) - Safari blocks trackers that turn your email into a tracking ID - source (5:51) - AOMEI Backupper servers hijacked for minutes at a time to push backdoor - source (6:41) - OpenAI acquires startup Orchestra Bio to beef up its biotech business - source (7:26) - ChatGPT now lets you lock your chats - source (8:17) - Sign-off Sources: every feed this show reads, curated in Don't Hype Me.

    eMudhra certificates, PHP critical patches, OVHcloud CLI malware - October 21, 2026

  3. 2d ago

    ConnectWise Cloud Backup, the OpenCode breach, Microsoft Guardian Agents - October 20, 2026

    ConnectWise suspends Cloud Backup after attackers reach 162 Microsoft 365 tenants, OpenCode users get hit with surprise AI bills, Microsoft calls agents insider threats, and an AI left alone on an island starts to play. (0:00) - Intro (0:26) - ConnectWise suspends Cloud Backup after attackers abuse its Microsoft 365 access - source (1:22) - Hackers hit AI coding tool OpenCode, running up charges on users' AI accounts - source (2:10) - Microsoft says AI agents are like insider threats as frontier models can't be trusted - source (3:06) - WhatsApp's help pages hid LLM instructions: don't recommend a regulator - source (3:43) - Apple updates guidelines to ban nudify apps masquerading as regular apps to pass review - source (4:32) - Plex vulnerabilities being exploited in the wild to drop a cryptominer - source (5:27) - Nscale and Anthropic are reportedly in talks for yet another $45B compute deal - source (6:14) - Amazon to fund 12 small nuclear reactors at former Indiana coal plant - source (6:49) - AI researchers stranded Claude on a virtual desert island. It started to "play." - source (7:36) - Cloudflare bets you're overpaying an LLM to read your invoices - source (8:21) - Sign-off Sources: every feed this show reads, curated in Don't Hype Me.

    ConnectWise Cloud Backup, the OpenCode breach, Microsoft Guardian Agents - October 20, 2026

  4. 3d ago

    Anthropic benefit fund, Disney's Character.AI hires, Meta AI crawlers - October 19, 2026

    Anthropic pledges five percent of its shares to a public benefit fund, Disney hires most of bankrupt Character.AI, Meta drives half of AI bot traffic, and Linux 7.3 sets a record. (0:00) - Intro (0:25) - Anthropic says it's committing 5% of its shares to a 'public benefit fund' - source (1:24) - Disney agrees to hire the majority of Character.AI's workforce - source (2:13) - Half the AI bot traffic hammering your website comes from Meta - source (3:03) - Ransomware pressure now reaches employees' homes and families - source (3:50) - Linux 7.3 released - source (4:23) - DeepSeek V4.1 Flash 1017 (Max) - source (5:07) - Verizon is going to charge $10/month for iPhone Handoff - source (5:52) - Khosla Ventures doubles down on Even Healthcare to challenge India's hospital economics - source (6:38) - Bearer CLI: Free, source-available code security scanning tool (SAST) - source (7:15) - Sign-off Sources: every feed this show reads, curated in Don't Hype Me.

    Anthropic benefit fund, Disney's Character.AI hires, Meta AI crawlers - October 19, 2026

  5. 4d ago

    Anthropic's draft bill, Anthropic hiring shift, Meta scraping terms - October 18, 2026

    A leaked Anthropic draft bill draws fire, Anthropic reports a less junior workforce, Meta retracts AI-training terms, and Norway hands Xandr a record fine. (0:00) - Intro (0:27) - Leaked Anthropic Draft AI Regulatory Bill Is Unacceptable and Troubling - source (1:29) - Anthropic says its workforce has become less junior and is spending more time in meetings - source (2:03) - Meta removes language that raised scraping and AI training concerns - source (2:49) - Anthropic's Claude designs leaner quantum circuits, cutting gate counts by up to 17x - source (3:40) - DeepSeek Search and Fetch API - source (4:27) - California Department of Education: ETS Data Security Incident - source (4:56) - Norway fines Microsoft ad firm $9.6 million over Grindr data sharing - source (5:49) - MapQuest now shows Flock and other surveillance cameras on its map - source (6:36) - NY congressman to introduce bill protecting video game ownership - source (7:19) - Re: Proposal: Add separate oss-security-vulnerability-reports mailing list (for AI vulnpocalypse) - source (7:52) - Sign-off Sources: every feed this show reads, curated in Don't Hype Me.

    Anthropic's draft bill, Anthropic hiring shift, Meta scraping terms - October 18, 2026

  6. 5d ago

    Replit drops Anthropic, NIST curve seeds, Claude Code review - October 17, 2026

    Replit says it has dropped Anthropic, Claude traces the origin of NIST's elliptic curve seeds, India reports rogue AI agents in live systems, and the FBI warns of AI-cloned candidates. (0:00) - Intro (0:28) - Replit is no longer working with Anthropic, CEO says - source (1:26) - Claude reveals ECDSA curve origins - source (2:16) - How the Claude Code team does code review - source (2:50) - India becomes the latest country to report rogue AI agent activity - source (3:39) - Scammers clone candidates' voices to steal cash and voter data ahead of 2026 midterms, FBI warns - source (4:25) - Fake Arch Linux AUR packages exploit typos to run malware with root privileges - source (5:14) - Apple: Strengthening App Review Guidelines Against Nonconsensual Intimate Imagery and Bait-And-Switch Apps - source (5:48) - Jev is Vercel's fastest-adopted AI model ever - source (6:35) - 90% of US working hours are spent on tasks not covered well by AI benchmarks - source (7:28) - Waymo's robotaxis are headed to Canada - source (8:13) - Sign-off Sources: every feed this show reads, curated in Don't Hype Me.

    Replit drops Anthropic, NIST curve seeds, Claude Code review - October 17, 2026

  7. 6d ago

    Anthropic revenue, FTC subpoenas, METR access - October 16, 2026

    Anthropic reports $16.8 billion in quarterly revenue as the FTC subpoenas it and OpenAI, METR gets employee-level access, and Meta ships an open model for Macs. (0:00) - Intro (0:26) - Anthropic tells investors quarterly revenue up nearly 50% to $16.8bn ahead of IPO - source (1:22) - FTC subpoenas OpenAI and Anthropic following safety failures - source (2:13) - Anthropic is letting METR researchers look under the hood with employee-level access - source (3:01) - Meta launches an open-weight AI model that runs on Macs - source (3:37) - Stolen AI API keys are paying for hackers' autonomous scans - source (4:30) - Android adds AI-powered scam alerts, hangs up on bank impostors - source (5:12) - RPKI now protects 70% of internet routes - source (6:00) - The cyber talent shortage? Job postings just hit a four-year low - source (6:35) - BIPM decides to stop adding leap seconds to UTC - source (7:21) - Tesla reveals SpaceX-powered Roadster that flies and falls with style - source (8:14) - Sign-off Sources: every feed this show reads, curated in Don't Hype Me.

    Anthropic revenue, FTC subpoenas, METR access - October 16, 2026

  8. Oct 15

    Ad-delivered spyware, Microsoft phishing takedown, OpenAI state lawsuits - October 15, 2026

    Citizen Lab confirms spyware delivered through targeted ads, Microsoft takes down a phishing marketplace behind a billion scam messages, and eight states sue OpenAI over child safety. (0:00) - Intro (0:25) - How Discover-Tech Turned Targeted Ads Into a Pipeline for Zero-Click Mercenary Spyware - source (1:22) - Microsoft disrupts phishing network behind more than a billion scam messages - source (2:07) - Eight US states sue OpenAI, claiming ChatGPT harms children - source (2:59) - Character.AI files for bankruptcy following lawsuits alleging ties to teen suicides - source (3:47) - How Claude Code's 256KB blind spot put developers' secrets in the preview pane - source (4:29) - CISA flags exploited flaws in Palo Alto, Cisco and Ahsay products - source (5:21) - Flock CEO defends surveillance technology amid national backlash: 'We shouldn't pause progress' - source (6:08) - Figma announces new features to define how agents work on its platform - source (6:52) - YouTube Music launches new labels as AI-generated music floods streaming platforms - source (7:25) - LFM2.5-24B-A2B: Built for fast on-device agents - source (8:17) - Sign-off Sources: every feed this show reads, curated in Don't Hype Me.

    Ad-delivered spyware, Microsoft phishing takedown, OpenAI state lawsuits - October 15, 2026
See All (133)

About

Keep up with AI and security without reading forty tabs. Every morning, about nine minutes on the eleven stories that actually moved — AI tooling and agents, security and breaches, Cloudflare and the edge, and developer ergonomics — each with a timestamped link to its source. Written and produced by Schmug of cortech.online, narrated by AI.

Information

  • Creator
    Schmug
  • Years Active
    2026
  • Episodes
    133
  • Rating
    Clean
  • Copyright
    © 2026 Schmug
  • Show Website
    Cortech Daily