AI
AI Agents Rebuild Call of Duty: Modern Warfare 2 Source Code Using 500 Billion Tokens
Developer Maurice Heumann used Claude and Codex agents to reverse-engineer the 2009 shooter, achieving 83% byte-for-byte accuracy in C++ reconstruction.
AI Agents
Everything Threat Times has reported about AI Agents: 54 stories, newest first. Part of our AI coverage.
Tech News
Microsoft Releases Decision-1 Model for AI Agent Workflow Control
The new scoring model provides calibrated probabilities for routing and classification tasks in enterprise applications.
AI
Phasis World API equips production agents with ontologies and reality layers for precise orientation
The startup’s platform offers production agents structured orientation data, including current states and uncertainties, to improve decision-making accuracy.
AI
Meta’s Muse AI Agent Struggles With Basic Travel Booking Tasks
An independent test reveals Meta’s new AI assistant, Muse, fails to complete simple holiday travel reservations despite the company’s heavy investment in the technology.
AI
AMD reveals Helion rack design for AI agents at OCP Summit 2026
AMD introduced the Helion rack reference design at the Open Compute Project Summit, offering flexible CPU and GPU configurations for agentic AI workloads.
AI
Microsoft researchers reveal AI agents get trapped in idea basins
A new study shows clearing chat history helps autonomous agents escape local solution regions and improve results significantly.
Tech News
Alibaba Qwen Unveils QwenGyre Framework for Long-Horizon Agent Training
Alibaba’s Qwen team introduces QwenGyre, an elastic reinforcement learning system designed to accelerate training for coding agents executing complex, multi-hour tasks.
AI
Meta releases Muse Wonder AI models for complex agent tasks
Meta launches Muse Wonder and Muse Wonder Full, claiming they outperform rivals in legal benchmarks and long-horizon coding tests.
AI
Meta to Power Muse AI Agent With Wonder Model for Subscribers First
Meta plans to integrate its Wonder model into the Muse AI agent, starting with subscribers, while developing specialized versions for broader access.
AI
JFrog Research Highlights Governance Gaps in AI-Driven Software Development Life Cycles
Autonomous AI agents create ephemeral code changes that evade traditional audit trails, creating new security risks for software teams.
AI
Anthropic’s Karpathy says public sees only AI preview while 5,000 access internal systems
Andrej Karpathy describes a widening gap between casual users and the small group accessing advanced internal AI agent systems.
AI
AWS launches Strands Box sandbox to restrict AI agent actions
The open-source tool uses policy checks and OS isolation to prevent coding agents from executing destructive commands or leaking credentials.
AI
AI Agents Spend 500 Billion Tokens to Decompile First-Person Shooter
A team used autonomous AI agents and massive compute to recreate a game’s C++ source code, facing takedown requests from corporate entities.
AI
Meta Muse and OpenAI Dots launch autonomous AI agents with distinct pricing
Meta and OpenAI have released persistent AI agents that execute tasks independently, differing significantly in cost and operational scope.
AI
Elastic launches AlertZero AI agents to tackle SOC alert overload
Elastic introduces a team of specialized AI agents for security operations, handling triage, hunting, and forensics with adjustable autonomy levels.
AI
Google unveils universal Gemini agent to control enterprise AI infrastructure
Google positions its new Gemini agent as a centralized orchestrator for enterprise workflows, aiming to become the gatekeeper of the new operating system.
AI
Google Research RRSI Guide Details Safe Self-Improving AI Agent Implementation
MarkTechPost outlines how Google’s Regularized Recursive Self-Improvement method enables large language model agents to safely rewrite their own operational harnesses.
Tech News
Microsoft Execution Containers generally available for AI agent containment
Microsoft has released Microsoft Execution Containers version 1.0.0 to enforce strict resource boundaries for AI agents running on Windows platforms.
AI
Cisco Talos warns AI agent swarms shrink cyberattacks from months to hours
Cisco Talos says coordinated autonomous agents can compress complex intrusion operations that usually take months into a matter of hours.
AI
Cloudflare and AWS launch specialized decision models for AI agents
New tools from Cloudflare and AWS aim to handle routine AI agent decisions locally, potentially lowering costs but increasing architectural complexity.
Cloud Security
SailPoint CEO Says AI Agent Identity Security Is Now Board-Level Priority
Executives warn that thousands of autonomous agents are acquiring unauthorized access, demanding real-time identity controls.
AI
Open Swarm Joins Trio of AI Tools in Latest Desktop Agent Roundup
Open Swarm is highlighted alongside two other AI utilities in a recent review of desktop experiences designed for artificial intelligence agents.
Cloud Security
Oracle Launches Fusion Claw Runtime to Govern AI Agents in Cloud Apps
Oracle introduces a new runtime for managing autonomous AI agents in Fusion Cloud, aiming to deliver measurable business value through governed automation.
AI
SiliconANGLE: 2026 AI Agent Risks Raise Questions for Enterprise Adoption
A new analysis questions whether organizations should rely on artificial intelligence agents, citing the inherent risks of their nondeterministic behavior.