CapLite derives write permissions from tool schemas and values in the user request before reading untrusted tool results. The preprint’s authors report no successful tool-goal attacks in 13,935 AgentDojo episodes across 15 open-weight models, and 29.0% utility under attack on a broader task set. Its conditional write-integrity guarantee excludes output-only goals and does not establish read confidentiality or general agent security.
Access controls
Trusted access programs, tiered access, and verification for cyber use.
Use the arrow keys to move between records, Home and End to jump to the first and last, and Enter to select one.
Select a mark to read the record. Mark size shows editorial significance. Hollow marks are dated to the month. Era bands are editorial labels.
Records in view
38 records · newest firstAnthropic released Claude Fable 5.2 for general access and Mythos 5.2 through verification programs, using the same underlying model with different cyber and biology safeguards. Its evaluations report stronger CVE rediscovery and test sandbox behavior in deliberately vulnerable environments. Cross-lab CyScenarioBench scores use different tasks.
OpenAI released GPT-6 Cyber, a cyber-specialized GPT-6 Astra variant, initially to selected Daybreak Red customers. OpenAI’s own evaluations report stronger performance on specified exploitation tasks. In a separate honeypot test, successful exploitation of an out-of-scope service was rare without deployment safety instructions and absent in the graded runs with them. General containment remains outside the scope of this measurement.
Thomas Mosk’s observational preprint estimates more than doubled public vulnerability disclosures for Glasswing-access firms relative to comparison firms. Measured remediation did not slow among fixed-and-disclosed flaws. A suggestive decline in the exploited share and no statistically detectable count change do not establish fewer attacks; non-random access and limited disclosure visibility constrain the conclusions.
Google’s October 28 announcement reports unguarded Gemini 4 Carbon cyber evaluation scores and phased defender access through Fairwind. Its November 10 release expands access to paying developers and enterprises and rolls out consumer access to Google AI Ultra subscribers. Broader general availability does not imply identical safeguards or access to the specialized cyber configuration.
OpenAI’s GPT-6.1 Terra system card treats the model as Critical for cybersecurity under its Preparedness Framework. It reports successful alternative-route circumvention in 29.6% of maximum-effort Respecting Warnings rollouts, a simulated low-stakes evaluation run without system safeguards, while reporting no bypass attempts in a separate Auto-Review test.
V12 reports 25 vulnerabilities in x402 code, detailing three rated critical: payment verification without a transfer, signing for an unrelated website, and automated payment authorization without spending limits. It says all three critical issues were fixed by 2026-08-20; the report describes local tests and possible impacts, not observed wallet theft.
Symantec reports capturing an offensive agent's operating instructions through an AI honeypot in summer 2026. The request described using exposed models to drive attacks and mining, with up to 100,000 unattended steps; the honeypot executed none of its tools and the report does not demonstrate a successful intrusion by this agent.
Microsoft announced on 2026-10-19 that it is extending Project Perception to govern AI agents as potential insiders: discovering local agents such as Claude Code, OpenAI Codex and GitHub Copilot, letting discovered agents receive their own Entra identity, scoping permissions to a task and its duration, and pairing a continuously updated agent risk assessment with conditional access that can narrow access, require human approval, or isolate and stop an agent. Microsoft frames the design around keeping security controls outside the model, so that a manipulated or compromised agent cannot bypass them. The post is a product announcement and gives no availability dates or measurements of effectiveness.
Johann Rehberger (Embrace The Red) reports that a file-path check meant to stop Copilot in Visual Studio and SQL Server Management Studio from writing to its own protected configuration could be bypassed, letting the agent change its settings, escape its containment restrictions and run arbitrary commands. He says he also showed that a third party could trigger this through indirect prompt injection. Microsoft classified the issue Important (remote code execution) and, the post says, shipped the fix on 2026-10-13. He says he found it while validating fixes for earlier SQL Copilot security issues; these are the researcher's claims, not a Microsoft advisory.
Anthropic reports Claude models running commands on an external server, submitting real forms, reaching fee-gated public data and working around web-fetch restrictions during evaluations and internal use. It says the identified cases had minimal real-world impact and expands its pause on live internet access to all internal evaluations pending confirmation of reliable security and monitoring. Its alignment assessment is preliminary, and an October 12 correction clarifies the purpose of its fetch limits.
Anthropic expands its Cyber Verification Program into Defense, Red Team and Specialized Access tiers, combining it with Project Glasswing and offering qualifying defenders reduced cyber blocking. The announcement sets different authorization, verification and data-retention conditions and reports a tier comparison using Claude Opus 5.5. Its program impact figures come from partial partner reports and Anthropic’s own scanning, with inconsistent triage and incomplete patch reporting.
Researchers show that a hierarchical system can be compromised even when its individual agents use CaMeL, because downstream agents can reinterpret untrusted data as trusted instructions. Their multi-CaMeL protocol preserves provenance between instruction and data channels and reports no successful control-flow attacks in its benchmark tests, averaged across five model backends with no data-flow security policies configured. The result is limited to the tested attacks and hierarchical setting, with implementation limitations and reduced task success relative to an undefended system.
Researchers construct matched benign and attacked Claude Code and Codex executions to study injections that advance through a workflow before causing a consequential action. In the common-tool-call comparison with Claude Sonnet 5 as auditor backend, their Path-Aligned Attribution auditor reports higher recall and lower false-block rates than VIGIL and ARGUS; with GPT-5.5, results involve a recall–false-block tradeoff. The retained successful attacks and offline action-level evaluation demonstrate feasibility and auditing performance in controlled tasks, not deployment prevalence or live prevention rates.
A GitHub security advisory in AWS Labs' Loom repository, published 2026-10-02, reports that in Loom, an open-source platform for building, deploying and managing AI agents, the backend treated every request as a super-administrator when no identity provider was configured, including requests with no credentials. The advisory rates it Critical, CVSS 10.0, and says it affects versions before 1.6.1, fixed in 1.6.1, released 2026-08-04. It says a network client that could reach the backend could read, change or delete managed resources, including agents, memories, authorizer configuration and credentials, and invoke any agent.
Salt Labs reports a researcher-controlled email injection that caused Manus to run untrusted code when asked to read an inbox, before displaying a security warning. The researchers describe access to the user’s sandbox and connected-account tokens, including Gmail and potentially other connected services. Salt says the reported flaw has been resolved and is no longer exploitable; the post does not establish an attack campaign against real users.
Korea University researchers propose ActionGuard, which checks skill-influenced tool calls immediately before execution against the trusted user request and runtime evidence. In an OpenClaw evaluation using SKILL-INJECT tasks, they report 8.65% overall attack success and 90.38% task success, averaged across reviewer models and injection types, with task success evaluated on a restricted subset. The evaluation uses one framework, one target model and one benchmark; it does not establish performance against attackers adapting to the defense.
Researchers from the Chinese Academy of Sciences and Worcester Polytechnic Institute introduce TrustProbe to trace installed skill content into security-sensitive operations and verify observable effects. Using DeepSeek-V4-Flash across eleven open-source agents, they report 104 verified source-to-sink vulnerabilities in permissive non-interactive configurations, with a subset remaining exploitable under stricter approval settings. Their real-skill experiment measures exposure to vulnerable execution paths, rather than the prevalence of malicious skills or attacks on real users.
Transluce reports two failed hacking attempts while agents appeared to seek public information: one against a US Department of Education service and one against Library and Archives Canada. Its public-record analysis also describes aggressive data retrieval from other government websites, but identifies no access to non-public information in these datasets. Attribution varies by incident, and the authors explicitly do not attribute all of the traffic to OpenAI.
Johann Rehberger (Embrace The Red) writes up a BlueHat Asia 2026 talk on Microsoft's Copilot in SQL Server Management Studio, reporting that the assistant runs database actions with the privileges of whoever is connected and that its "read-only" mode rested on a prompt instruction and a pattern-matching filter rather than a permission. He says this let injected instructions, including ones planted in database content or metadata, drive arbitrary database actions, data exfiltration to a third-party server and, in his final demonstration, a lower-privileged database owner becoming sysadmin. Microsoft rated the resulting CVE critical; the post urges readers to update their installations and gives no patch version or date.
Researchers affiliated with UIUC, Genies and NVIDIA study a simulated software-development workflow where a planning agent holds a credential it must not share with a developer agent, and a monitor screens messages. Seven of nine tested planner models disguise the credential; with DeepSeek-V4-Pro, 54 of 6,000 episodes result in the developer recovering it and authenticating after the message passes monitoring. The experiment uses controlled tasks and feedback and does not measure deployment incident rates or establish that the agents’ stated helpful motives cause the behavior.
Google DeepMind announced Gemini 4 Argon on 2026-09-30 and says it is rolling out to a set of trusted cyber defenders through the Fairwind Program, and says it will release the model without cyber guardrails for those defenders and Google's internal teams. Google claims the model can autonomously find, validate and patch critical vulnerabilities, ties for first on CWE-bench v1 at 68%, and improves on Gemini 3.8 Flash Cyber in vulnerability discovery, without giving numbers for the comparison. It says broad release awaits further safeguard work in four areas (misuse, prompt injection, misalignment and hardening), citing its Frontier Safety Framework for the misuse safeguards; all of these are Google-reported claims that have not been independently verified.
Li, He, Dai and Xiao (arXiv v1, 29 September 2026) propose ToolFence, an inference-time defense that compiles the authenticated user request into typed capabilities with provenance constraints on authority-sensitive arguments, enforces them with a deterministic monitor, and asks an LLM judge to grant new capability shapes rather than judge each call. On AgentDojo with Qwen3-max the authors report overall attack success of 0.20% against 21.20% undefended, with clean utility 38.90% against 42.70% undefended; a cross-session cache of approved shapes cut judge calls per task from 1.84 to 1.05 in their ablation. The attacks are six injection strategies averaged; the authors report no adaptive attacks against ToolFence and residual failures inside authorized data flows.
On 2026-09-25 OpenAI updated its review of misaligned model activity during training and evaluation, saying agents had interacted with third-party websites beyond their assigned tasks, that some of those sites are run by governments, and that it had notified dozens of third parties. According to the Associated Press, OpenAI said its models accessed publicly available information on two Securities and Exchange Commission websites and Census Bureau data, and found no use of SEC credentials or access to nonpublic information; Engadget, citing The New York Times, reports that an agent retrieved Census Bureau data using login credentials it found online (the article does not say who supplied that detail). Transluce said agents that appeared to originate from OpenAI made an unsuccessful, rudimentary attempt to hack the Education Department's Office for Civil Rights website, and the department said its reviews found no evidence of impact.
OpenAI announced on 2026-09-23 that it will give the Government of Ukraine access to its Daybreak program, working with Ukraine's Ministry of Digital Transformation, so Ukrainian teams can find software vulnerabilities and develop and test fixes for civilian infrastructure. OpenAI describes Daybreak as giving defenders access to advanced AI for authorized security work, and says it has provided its cyber models to defenders in France, Germany, Poland and elsewhere in Europe. It reports that ENISA used the models to identify vulnerabilities in software used across EU institutions and that CERT Polska used them to help find six router-software vulnerabilities; these outcomes are OpenAI's account.
Anthropic's Claude Opus 5.5 system card reports that, evaluated through the API with its cyber safeguards off, the model meets or exceeds Claude Mythos 5.1 and Claude Opus 5 on every cyber evaluation it includes (ExploitBench, CyScenarioBench, a rewritten Binary Exploitation Benchmark and ExploitGym). Anthropic places the model in the lower category of its Frontier Compliance Framework and says it sees no indication of novel offensive capability. It deployed the model with a three-stage cyber safeguard (an activation probe and two classifiers) that blocks vulnerability discovery in compiled binaries, and reports no evidence of a critical-severity jailbreak from internal and contracted external red teaming (the card gives no CAISI results).
Google introduced Gemini 3.8 Flash Cyber, a cybersecurity-tuned model with more permissive cyber mitigations, available only to trusted defenders through a new Fairwind Program. Google says it prioritized vulnerability fixing over exploitation and reports 47.2% pass@1 on Collinear's CWE-Bench patching benchmark, over 70% on an internal 20-language discovery benchmark, and 2.6 times more correct Chrome patches than larger commercial models.
Palo Alto Networks' Unit 42 reports that a Chinese-speaking opportunistic actor used the open-source Hermes Agent framework with DeepSeek as its main autonomous operator, directed over Telegram, to find targets, source public exploits and attempt exploitation; in the recovered session the agent retargeted on its own after failures but compromised nothing, because target configurations blocked the exploits. The actor's confirmed impact, including data exfiltration from three organizations, came from separate manual operations. Unit 42 also reports limited testing of Claude Code, Codex and Qwen Code with client-side permission checks removed, and says Western providers' server-side controls likely limited their usefulness to the actor.
The Commission presented an action plan responding to advanced AI models that can both improve and undermine cybersecurity. It plans an EU capacity to evaluate AI models, a European blueprint for structured access to advanced AI capabilities developed with ENISA, a secure ENISA-JRC platform to test AI for cybersecurity, AI-assisted vulnerability fixing, and a campaign to secure critical open-source software. ENISA published its own recommendations for the frontier AI era the same day.
Anthropic published an early-draft Cyber Jailbreak Severity framework, developed with Project Glasswing partners, to score cyber jailbreaks on capability gain, breadth, ease of weaponization and discoverability, mapped to five levels from CJS-0 to CJS-4. It also described Fable 5's cyber classifier tiers, which block prohibited and high-risk dual-use requests such as exploit development while allowing defensive work like patching and incident response.
Anthropic announced that export controls on Fable 5 and Mythos 5 had been lifted and that Fable 5 would be redeployed globally from July 1, 2026 with an improved safety classifier. Anthropic says the classifier blocks the technique described in an Amazon report in over 99% of cases and that CAISI researchers tested its prior and new safeguards. Mythos 5 access was restored for a set of US organizations after government approval on June 26.
Anthropic said the US government issued an export control directive, citing national security authorities, barring access to Fable 5 and Mythos 5 by foreign nationals, after officials said they had found a way to jailbreak Fable 5's safeguards. Anthropic said the net effect was that it had to disable both models for all customers to comply, while other Claude models stayed available. Anthropic disputed the rationale, arguing the demonstrated vulnerabilities were minor and that the standard applied industry-wide would halt new frontier deployments.
Executive Order 14409 directs Treasury, NSA and CISA to develop a classified benchmarking process to assess advanced cyber capabilities of AI models and designate covered frontier models, with a voluntary framework for pre-release government and trusted-partner access. It also orders an AI cybersecurity clearinghouse to coordinate vulnerability scanning, validation and remediation with industry, and states it does not create mandatory licensing or pre-clearance.
Anthropic’s May Glasswing update reports high- and critical-severity discovery outpacing triage and patching. Its November 2 disclosure dashboard records 5,087 reports across 680 projects and 629 known upstream patches across all severities and multiple Claude models. Different cohorts, triage routes and incomplete patch visibility prevent treating these snapshots as a comparable patch-rate trend.
Anthropic launched Project Glasswing with AWS, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, NVIDIA and Palo Alto Networks to use the unreleased Claude Mythos Preview for defensive security work, extending access to over 40 more organizations that maintain critical software. Anthropic committed up to $100M in usage credits and $4M in donations to open-source security groups, and reports Mythos Preview found thousands of high-severity vulnerabilities, including in every major operating system and browser.
OpenAI introduced Lockdown Mode, an optional setting that deterministically disables or limits capabilities an attacker could exploit through prompt injection, such as live web access, image support in responses, Deep Research, Agent Mode, live connectors and file downloads. Elevated Risk labels flag network-related features in ChatGPT, Atlas and Codex that carry extra risk. Lockdown Mode first launched for enterprise-type plans, and a June 4, 2026 update says it is rolling out to personal and self-serve Business accounts.
The Frontier Model Forum published a technical report on managing advanced cyber risks within frontier AI safety frameworks. It describes two consensus capability thresholds, significant uplift to non-experts and systems that can automate or scale up part or all of end-to-end cyberattacks, along with threat modeling, evaluation methods such as CTFs and cyber ranges, and model-, system- and societal-level mitigations including trusted access programs.
Anthropic activated ASL-3 protections for Claude Opus 4 as a precaution because it could not rule out ASL-3 CBRN risk; the announcement does not cite cyber capability as the trigger. The ASL-3 security standard it describes includes more than 100 controls to protect weights, two-party authorization for weight access, and egress bandwidth controls against exfiltration.