Market & Regulation
New Worm Targets AI Infrastructure in Supply Chain
WIRED AIhigh confidence
Why it matters
- →The worm exploits AI development pipelines, a growing target for attackers.
- →Its stealthy nature makes it difficult to detect, posing a significant security challenge.
- →Highlights the need for enhanced security measures in AI software development.
CrowdStrike has identified a new worm that targets AI infrastructure, posing a significant threat to the AI software supply chain. The worm conducts reconnaissance, steals access credentials, and can even destroy data, all while blending in with legitimate AI coding activities. This makes detection challenging, as the worm's behavior closely resembles normal operations in AI development environments. The discovery highlights the need for improved security measures as AI tools become more integrated into software development processes.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1Red-teaming AI agent networks reveals new vulnerabilities
Microsoft Research · April 30, 2026 · Background
- 2Government Scrutiny on AI Supply-Chain Risks
The AI Daily Brief · May 2, 2026 · Background
- 3Malicious Software Found on Hugging Face Platform
AI News · May 12, 2026 · Background
- 4AI Transforms Cyber Threat Landscape, Report Finds
Anthropic · June 3, 2026 · Background
- 5Meta AI hack exposes security vulnerabilities
MIT Technology Review AI · June 5, 2026 · Background
- 6New Platform FLARE-AI Tracks AI Misbehavior
WIRED AI · July 1, 2026 · Background
- 7MIT Develops Method to Detect Harmful AI Models
MIT News AI · July 13, 2026 · Background
- 8Defenders Use Prompt Injections to Stop AI Attacks
WIRED AI · July 18, 2026 · Background
What happened next
- 9NVIDIA and CrowdStrike Launch Agentic Cybersecurity System
NVIDIA Blog · September 1, 2026 · Background
- 10OpenAI publishes Hugging Face hack postmortem
Fireship · September 2, 2026 · Background
- 11OpenAI Agents Implicated in RubyGems Attack
The Verge AI · September 12, 2026 · Background
- 12AI Agents as Malware Distribution Channels
AI News · September 23, 2026 · Background
- 13Massive AI-Fueled Cyberattack Targets 100 Companies
AI Explained · September 24, 2026 · Background
More from WIRED AI
© WIRED AIGeneral AIagents
Meta's Muse Mascot Sparks Privacy and Age Debate
Meta’s AI agent Muse is wrapped in a deliberately cute aesthetic that experts warn disarms adult users into passive compliance with surveillance. The mascot, Jolly, resembles a children’s toy despite the app being strictly 18+, prompting backlash from youth advocates who fear it blurs age boundaries. This design choice serves a strategic purpose: lowering psychological barriers to trust for a company facing significant reputational damage from past privacy scandals and addiction lawsuits. As Meta prepares to launch a physical Tamagotchi-style device, the tension between delightful UX and data extraction becomes the central conflict of AI adoption.
WIRED AI
© WIRED AIMarket & Regulationbusiness
Appeals Court Upholds Pentagon Ban on Anthropic
The DC Circuit Court has upheld the Pentagon’s designation of Anthropic as a supply-chain risk, effectively banning Claude from military systems. The ruling validates the Trump administration's stance that Anthropic’s refusal to allow autonomous weapons deployment constitutes a national security threat. This legal win allows the DoD to continue replacing Claude with competitors like Grok and Gemini while Anthropic faces potential revenue loss ahead of its IPO. The decision marks a significant escalation in government leverage over AI safety constraints, setting a precedent for how model restrictions impact commercial viability.
WIRED AI
© WIRED AIInvestment · $1 billion
Market & Regulationagents
Instinct AI raises $1B on agent hype
Instinct is a stealth-mode startup that has reportedly secured talks to raise $1 billion, pushing its valuation to $10 billion. The company’s pitch relies on a deceptively simple form factor: an autonomous agent that lives inside iMessage and WhatsApp rather than a dedicated app. While early user experiences highlight genuine utility in booking restaurants and navigating complex flight refunds, the product is riddled with security risks, aggressive API usage, and opaque data policies. This massive valuation signals that investors are betting heavily on the consumer agent narrative despite significant technical and trust hurdles.
WIRED AI
More in Market & Regulation
Market & Regulationother
Lisbon startup insures AI agents
Insurance for autonomous systems is a nascent market, and this Lisbon startup is positioning itself at the frontier. By targeting liability for AI agents specifically, they address a critical gap as these systems move from experimental to operational in enterprise workflows. This signals that the industry is beginning to treat agent-driven actions with the same risk management rigor as traditional software or human labor. It’s an early bet on the legal and financial infrastructure required for widespread agent adoption.
Sifted
© TechCrunch AIMarket & Regulationother
Anthropic CEO to meet Trump at White House
The meeting between Anthropic's Dario Amodei and President Trump marks a significant pivot in the political landscape of AI safety. Despite recent public clashes over development speed and safety guardrails, this dinner signals a potential thaw in relations that could influence future regulation. The Pentagon's prior designation of Anthropic as a supply-chain risk adds complexity to this engagement. This event underscores how top AI executives are becoming central figures in shaping national policy. The shift from adversarial stances to direct dialogue suggests a new era of industry-government collaboration. Amodei’s presence at the White House elevates the profile of safety concerns in high-level political discourse. This encounter could redefine the boundaries of acceptable AI development under federal oversight.
TechCrunch AI
© TechCrunch AIMarket & Regulationbusiness
Google tests direct Flipkart checkout in Gemini
Google is moving AI from recommendation to transaction by testing a native 'Buy' button for Walmart-owned Flipkart within Gemini and AI Mode. This test allows users to complete purchases without leaving the interface, marking a tangible step toward agentic commerce ahead of India's festive season. Unlike Google's Universal Commerce Protocol which often redirects or uses hosted checkout, this integration appears to bridge directly into Flipkart's branded flow. It signals that search giants are prioritizing closed-loop revenue capture over open standards in key markets.
TechCrunch AI