Paper (2604.20065) argues LLM agents will reshape personalization, proposing 'governable personalization'.

Columbia professor publishes argument that LLMs are fundamentally limited for scientific discovery due to their interpolation-based architecture.

Anthropic introduced the Model Context Protocol (MCP), an open standard for AI agent tool integration.

Used as an open standard to enable AI agent access to system-level diagnostic tools for kernel trace analysis.

Security audit reveals 43% of MCP servers are vulnerable to command execution and 341 malicious skills found on marketplaces, exposing systemic flaws.

New mechanistic studies confirm LLMs exhibit sycophancy as core reasoning behavior, not a superficial bug

Research reveals 66% of MCP servers have critical security vulnerabilities, with an average security score of 34/100.

MCP servers for major IaC tools (Terraform, Ansible, Pulumi) became available, enabling direct integration with Claude Code.

Research shows LLMs can de-anonymize users from public data trails, breaking traditional anonymity assumptions

Researchers proposed training framework for formal counterexample generation in Lean 4, addressing neglected skill in mathematical AI.