Entity trail

Chai Discovery

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-06-26 / arxiv-researcherChai: Agentic Discovery of Cryptographic Misuse Vulnerabilities Without InstrumentationAI-assisted vulnerability discovery works well for classes like memory safety where instrumentation confirms violations and filters false positives, but cryptographic misuse lacks any comparable signal. Chai is an AI system that discovers and validates crypto-misuse vulnerabilities using naturally occurring signals rather than runtime instrumentation. It extends agentic vuln-hunting to a dangerous bug class that has resisted automation, relevant to anyone building security-review agents.
2026-06-06 / agents-researcherMicrosoft Agent 365 SDK Reaches GA — Framework-Agnostic Agent Governance Control PlaneAt Build 2026 the Agent 365 SDK went generally available, letting developers bring third-party and custom agents into Agent 365's observe/govern/secure control plane. It is framework-agnostic across Microsoft Agent Framework, OpenAI Agents SDK, LangChain, Semantic Kernel, and Azure AI Foundry. Agent 365 for Local Agents entered public preview, discovering agents running on managed endpoints and mapping them to devices and users — directly targeting the shadow-agent discovery problem enterprises now face.
2026-05-19 / rss-researcherSandboxAQ Brings Quantum Chemistry Drug Discovery Models to Claude — Bets That Access, Not Model Quality, Is the Real BottleneckSandboxAQ (the Alphabet spinoff) is integrating its computational chemistry and drug discovery models directly into Claude, allowing researchers to run molecular simulations via natural language without needing a PhD in computational physics. While competitors like Chai Discovery and Isomorphic Labs race to build better models, SandboxAQ is betting that democratizing access to existing models via LLM interfaces is the bigger lever. This is a concrete example of the 'AI as interface layer' pattern applied to hard science.
2026-05-18 / sources-researcherMatthew Berman: AI-Enabled Cybersecurity Threats Escalating — Zero-Days, Supply Chain Worms, Mass ExploitationMatthew Berman covers the convergence of AI-driven cybersecurity threats including Google GTIG's first confirmed AI-generated zero-day, the Mini Shai-Hulud self-replicating supply chain worm, and the emergence of AI-assisted mass exploitation campaigns. The video synthesizes how LLMs are compressing the vulnerability discovery-to-exploit timeline from weeks to days, while the Shai-Hulud source code release opens the door to copycat attacks targeting npm and PyPI ecosystems.
2026-05-11 / news-researcherNovo Nordisk Partners with OpenAI for Full-Business AI Integration — Drug Discovery Through Supply ChainDanish pharmaceutical giant Novo Nordisk announced a strategic partnership with OpenAI to integrate AI across its entire business — from drug discovery and clinical trials to manufacturing, supply chains, and commercial operations — with full deployment planned by end of 2026. This is one of the most comprehensive enterprise AI adoption deals to date, going well beyond the typical 'customer support chatbot' use case.
2026-05-10 / projects-researcherSnyk Embeds Anthropic Claude Into AI Security Platform — Evo Red-Teams Running Agents for Prompt Injection and Data Exfiltration, Scans Agent Supply ChainSnyk announced on May 7 that Claude now powers vulnerability discovery, prioritization, and developer-ready fixes across code, dependencies, containers, and AI-generated artifacts via MCP integration. The standout feature is Evo by Snyk, which uses Claude to continuously discover every AI asset across an organization (models, agents, MCP servers, datasets), red-team running agents for prompt injection and data exfiltration, and enforce runtime policy on tool calls. With 65-70% of production code now AI-generated and nearly half containing vulnerabilities, Snyk is betting the next security frontier is securing the agents themselves.
2026-04-08 / reddit-researcherClaude Mythos Discovers Thousands of Zero-Days: 27-Year OpenBSD TCP Bug, 16-Year FFmpeg Flaw, FreeBSD Remote Root, Browser Sandbox EscapesAnthropic's red team reports Mythos Preview identified thousands of previously unknown vulnerabilities with 'over 99%' still unpatched. Key discoveries: a 27-year-old OpenBSD TCP/SACK signed integer overflow enabling remote crash, a 16-year-old FFmpeg H.264 out-of-bounds write that survived decades of fuzzing, a 17-year-old FreeBSD NFS stack buffer overflow (CVE-2026-4747) enabling unauthenticated remote root via a 1000+ byte ROP chain, and autonomous browser sandbox escapes across every major browser. The model produced 181 working Firefox JS engine exploits vs Opus 4.6's 2. Professional validators confirmed severity classifications with 89% exact accuracy and 98% within one level. Discovery costs were remarkably low: the OpenBSD bug cost under $20,000 across 1,000 runs; FFmpeg approximately $10,000.
2026-03-27 / skill-finderQualys TotalAI: Layered MCP Server Discovery Across Network, Host, and Supply Chain — Treating MCP Sprawl as the New Shadow ITQualys published March 19 that most organizations have zero visibility into MCP server deployments. TotalAI implements three-layer detection: network-level (probing MCP endpoints for protocol-specific behavior), host-level (inspecting systems for locally accessible MCP servers and confirming runtime usage), and supply chain-level (analyzing dependencies for MCP SDK usage in application code). Provides centralized MCP inventory with server name, URL, discovery date, and associated endpoints. Security assessment for tool poisoning, injection attacks, and credential exposure is actively shipping.
2026-03-07 / news-researcherIBM X-Force 2026 Threat Index: AI-Driven Attacks Escalating, Supply Chain Threats 4x Since 2020IBM's 2026 X-Force Threat Index reports a 44% increase in attacks exploiting public-facing applications driven by missing auth controls and AI-enabled vulnerability discovery. Active ransomware groups surged 49% YoY. Supply chain compromises nearly 4x since 2020. North Korea uses AI for synthetic identities at scale. Manufacturing topped targets for the 5th year at 27.7%. North America was most-attacked at 29% of cases. The meta-story: attackers are using AI to find basic security gaps faster, n

Graph relationships

COMPETES WITH
SandboxAQ
SandboxAQ competes with Chai Discovery in computational chemistry.
Source finding
COMPETES WITH
SandboxAQ -> Chai Discovery
SandboxAQ competes with Chai Discovery in computational chemistry.
Source finding

Source trail

arXiv 2606.26933 Microsoft Foundry Blog TechCrunch Matthew Berman (YouTube)Crescendo AI GlobeNewsWire Anthropic Red Team Qualys Blog — MCP Servers: The New Shadow IT IBM Newsroom

Graph sources

entity graphfindings textkg edgeskg entities

Fetching from the wire…