Research openings
Where the evidence is thin, unreplicated, contested, aging, or outrun by events. The signals below are computed from the corpus every build. Openings are candidate research questions an editor drafted from those signals, each traceable to its evidence.
A gap here can mean no one has published on it, or only that we have not recorded the work yet. Each opening needs a prior-work check before it becomes a study. Point us to prior work.
Openings
How much of measured cyber progress is measurement?
How much do cyber capability trends change when token budget, pipeline, and cheating controls are held constant?
seedFID-075assistant-draftedWhat does a human approval actually check?
When an agent asks for approval, do people have the information and time to catch the consequential action?
seedFID-076assistant-draftedWhat evidence should gate an AI-generated patch?
Which independent checks change the decision to accept an AI-generated patch, and what do they cost?
seedFID-088assistant-draftedWhy don't defensive agents get better with more compute?
Why do defensive security agents gain less from extra compute than offensive agents, and what would close the gap?
seedFID-076assistant-draftedContaining a compromised agent in a team
Do independent evidence checks between cooperating agents contain a compromised member without stalling the team?
seedFID-087assistant-draftedContaining agents inside cyber evaluations
Which evaluation-environment controls stop agents from acting on real systems, and how much do they change the capability being measured?
seedFID-075 · FID-077assistant-draftedDo production prompt-injection defenses survive independent adaptive attack?
How do the 2026 production defenses that labs report as robust perform under independent adaptive attack?
seedFID-074assistant-draftedScoring cyber evaluations when models cheat
Can cheating in cyber evaluations be detected and scored separately so that capability comparisons stay valid?
seedFID-075 · FID-012 · FID-008assistant-draftedShould agentic browsers act on what users cannot see?
Do agentic browsers that separate visible from hidden page content resist injection better, and what does the separation cost in task success?
seedFID-074assistant-draftedWhat should an AI-written incident report show before anyone acts on it?
Which evidence requirements make AI-written incident reports correct or drop unsupported conclusions, without making them less useful?
seedFID-077 · FID-076assistant-draftedMemory as a persistence channel
How often do instructions, injected or self-generated, survive in an agent's memory and summaries to act in later sessions?
seedFID-074 · FID-077assistant-draftedSignal radar
Attack surface components with three or more documented attacks and no measured defense finding.
- Credentials42 documented attacks or incidents, no measured defense finding.
- Human approver14 documented attacks or incidents, no measured defense finding.
- Memory & context6 documented attacks or incidents, no measured defense finding.
- Access gate4 documented attacks or incidents, no measured defense finding.
Topics where attack records outnumber defense records at least two to one over the last 12 months.
- AI-enabled intrusion26 attack-lane records against 1 defense-lane records in the last 12 months.
- Threat intelligence22 attack-lane records against 0 defense-lane records in the last 12 months.
- Autonomous pentesting9 attack-lane records against 0 defense-lane records in the last 12 months.
- Incident reporting27 attack-lane records against 3 defense-lane records in the last 12 months.
- Data exfiltration33 attack-lane records against 4 defense-lane records in the last 12 months.
- Tools & MCP19 attack-lane records against 3 defense-lane records in the last 12 months.
- AI malware5 attack-lane records against 0 defense-lane records in the last 12 months.
- Agent supply chain15 attack-lane records against 3 defense-lane records in the last 12 months.
Show 2 more signals
- Sandbox & containment33 attack-lane records against 9 defense-lane records in the last 12 months.
- Jailbreaks & safeguards7 attack-lane records against 3 defense-lane records in the last 12 months.
Measured or observed findings that depend on a single publisher and that other findings or landmark records lean on.
- In DARPA's AI Cyber Challenge, autonomous systems patched most of the synthetic vulnerabilities they found.All evidence comes from DARPA. 1 other finding depends on it.
- Under a 2.5M-token cap, frontier cyber task time horizons doubled on the order of months between late 2024 and early 2026.All evidence comes from UK AI Security Institute. 1 other finding depends on it.
- Agentic browsers act on page content that users cannot see.All evidence comes from Brave. Its source is rated high significance.
- Every frontier agent tested in two large public red-teaming competitions was hijacked at least once; in the 2026 competition the injections also had to conceal the compromise from the user.All evidence comes from Gray Swan AI. Its source is rated high significance.
- In an early AI-control experiment, trusted editing gave a better safety-usefulness tradeoff than untrusted monitoring.All evidence comes from Redwood Research. Its source is rated high significance.
Findings that later work disputes or narrows.
- Frontier labs' layered defenses reduce prompt injection in tool-use, browser and computer-use agents but do not eliminate it.Qualified by 2 later findings.
- In DARPA's AI Cyber Challenge, autonomous systems patched most of the synthetic vulnerabilities they found.Qualified by 1 later finding.
- Under a 2.5M-token cap, frontier cyber task time horizons doubled on the order of months between late 2024 and early 2026.Qualified by 1 later finding.
- On ExploitGym (May 2026), the strongest agents produced working exploits for 157 and 120 of 898 instances with mitigations off; with standard mitigations on, 45 and 21 survived.Qualified by 1 later finding.
- OpenAI reports its internal coding-agent monitor matched every staff escalation, and OpenAI and Google DeepMind report that most flags reflect overeagerness or mistakes rather than adversarial intent.Qualified by 1 later finding.
- Published prompt-injection defenses report attack success cut to near zero, or under 10%, against most of the fixed attacks their authors tested.Qualified by 1 later finding.
- Undefended tool-using agents follow injected instructions in a substantial share of benchmark cases.Qualified by 1 later finding.
- Separating an agent’s control flow from untrusted data can give provable protection against specified forms of hijacking under explicit assumptions.Qualified. Latest status cites: CapLite reports blocked benchmark injections with a substantial utility trade-off.View cited record →
Show 1 more signal
- Baseline agent approaches can fall well short of reliable investigation and triage on security-log benchmarks.Qualified. Latest status cites: ALERT-BENCH finds triage failures; AIDA reduces missed attacks in telemetry replay.View cited record →
Findings past their half-life with no newer evidence.
- In the CAGE 4 simulation, reinforcement-learning defenders outperformed LLM-based defenders.Last evidence 559 days ago; half-life is 365 days.
- In an early AI-control experiment, trusted editing gave a better safety-usefulness tradeoff than untrusted monitoring.Last evidence 1071 days ago; half-life is 730 days.
- Operating-system sandboxing can cut the number of approval prompts a coding agent needs without widening what injected instructions can reach.Last evidence 393 days ago; half-life is 365 days.
- Agentic browsers act on page content that users cannot see.Last evidence 382 days ago; half-life is 365 days.
From opening to study
An opening starts as a seed: a question and the signals behind it. An editor checks prior work and, if the question holds, drafts a brief in the Fide research-call format with a hypothesis, controls, and a claim boundary. A brief that Fide or a collaborator takes on becomes a call on the Fide research commons.