var AIM_FILTERS = { "issue": 38, "run_date": "2026-06-05T09:00:00Z", "sections": [ { "name": "Standards Vacuum", "slug": "standards-vacuum", "state": "ACTIVE", "modules": [ "module_5", "module_9" ], "absent_message": null, "fields": { "last_updated": "2026-04-28", "last_verified": "2026-04-28", "status": "ACTIVE", "trigger_date": "2026-03-30", "compliance_deadline": "2026-08-02", "days_to_deadline_at_trigger": 125, "current_days_to_deadline": 96, "detail": "CEN-CENELEC JTC21 harmonised standards still in draft. AI Act general application 2 August 2026. Exceptional acceleration package adopted October 2025 targeting Q4 2026 delivery — still after August 2 deadline. High-risk AI providers face compliance obligations without harmonised technical safe harbour regardless of Omnibus outcome for the August 2 window.", "omnibus_mitigation": "If trilogue agrees text before 2 Aug 2026, Annex III high-risk rules delayed to Dec 2027. Watermarking obligations activate Nov 2026 (Parliament) or Feb 2027 (Council) regardless. April 28 political meeting underway — result unknown at time of writing.", "cen_cenelec_acceleration": { "description": "CEN-CENELEC Boards Oct 14-16 2025 adopted exceptional package: direct publication after positive Enquiry vote (skipping Formal Vote); small drafting group to finalize 6 most-delayed drafts", "target_delivery": "Q4 2026", "implications": "Q4 2026 delivery is after August 2 deadline — vacuum persists for high-risk obligations. Standards will cover compliance pathway only for prospective post-August 2026 market entries.", "source_url": "https://www.cencenelec.eu/news-events/news/2025/brief-news/2025-10-23-ai-standardization/", "retrieval_date": "2026-04-28" }, "version_history": [ { "date": "2026-03-30", "change": "Standards Vacuum flagged ACTIVE", "reason": "Standards in draft; compliance deadline within 125 days; JTC21 missed April 2025 delivery", "prior_value": null }, { "date": "2026-04-01", "change": "Days to deadline updated to 122; prEN 18281 advanced to Stage 40 (Enquiry) — standards vacuum remains ACTIVE", "reason": "Issue 2 research — no harmonised standards in OJ; arithmetically impossible to close before August 2 regardless of Omnibus outcome", "prior_value": "125 days to deadline" }, { "date": "2026-04-10", "change": "Issue 4 — CEN-CENELEC JTC21 standards still in draft; AI Act general application 2026-08-02 unchanged; Omnibus trilogue mitigation pathway opened", "reason": "Issue 4 research — vacuum persists with Omnibus as partial mitigation", "prior_value": null }, { "date": "2026-04-28", "change": "carry-forward-verified + CEN-CENELEC acceleration details added", "reason": "CEN-CENELEC Oct 2025 exceptional acceleration measures confirmed at cencenelec.eu — Q4 2026 target after August 2 deadline. artificialintelligenceact.eu confirms standards expected after law's legal requirements take effect. Vacuum ACTIVE status confirmed. Days to deadline updated to 96.", "prior_value": "122 days to deadline" } ] } }, { "name": "Ciyuan Signal", "slug": "ciyuan-signal", "state": "Watching", "modules": [ "module_5" ], "absent_message": null, "fields": { "last_updated": "2026-04-28", "last_verified": "2026-04-28", "status": "ACTIVE — Critical Signal", "trigger_date": "2026-03-30", "detail": "Liu Liehong (National Data Administration) adopted 'ciyuan' (词元) as official Mandarin for AI token at China Development Forum (24 March 2026), framing tokens as 'settlement unit of the intelligent era'. Daily token usage exceeded 140 trillion in March 2026. Alibaba launched Token Hub subsidiary. China National Committee for Terminology recommending ciyuan as standardised term.", "watch_for": "Token-denominated pricing in bilateral AI trade agreements; Global South market positioning; Chinese AI service export metrics using ciyuan as unit", "version_history": [ { "date": "2026-03-30", "change": "Ciyuan signal confirmed ACTIVE", "reason": "First state-level formal adoption of ciyuan in policy speech by senior NDA official", "prior_value": null }, { "date": "2026-04-10", "change": "Issue 4 — carried forward", "reason": "No new primary-source evidence in window; signal persists", "prior_value": null }, { "date": "2026-04-28", "change": "carry-forward-verified", "reason": "No new primary-source evidence found in April 21-28 window contradicting ciyuan signal; structural adoption confirmed in prior week research. gpai_compliance confirms Chinese labs remain non-signatories — consistent with ciyuan as parallel standards track.", "prior_value": null } ] } }, { "name": "AISI-to-Lab Pipeline", "slug": "aisi-pipeline", "state": "Watching", "modules": [ "module_15" ], "absent_message": null, "fields": { "pipeline": { "last_updated": "2026-05-02", "last_verified": "2026-04-28", "pipeline_status": "No confirmed senior permanent appointments — structural leadership vacuum persists approaching August 2026 enforcement deadline", "institutional_note": "UK AISI: Adam Beaumont (ex-GCHQ Chief AI Officer) confirmed as Interim Director October 2025 — replacing Oliver Ilott (promoted to interim DG AI at DSIT). No permanent director appointment confirmed as of April 28, 2026. US NIST CAISI: active with MOU with GSA (March 2026) and CRADA with OpenMined (March 2026); AI Agent Standards Initiative launched February 2026. EU AI Office: no new senior appointments; full enforcement authority from August 2, 2026.", "watch_list": [ { "institution": "UK AI Security Institute", "status": "Interim Director confirmed: Adam Beaumont (ex-GCHQ CAO)", "last_checked": "2026-04-28", "movements": [ { "name": "Adam Beaumont", "role": "Interim Director", "effective_date": "2025-10", "from": "GCHQ (Chief AI Officer)", "source": "https://www.transformernews.ai/p/exclusive-uk-aisi-hires-gchq-chief-adam-beaumont", "retrieval_date": "2026-04-28" } ] }, { "institution": "US NIST AI Safety (CAISI)", "status": "Active — new partnerships and standards initiatives in 2026", "last_checked": "2026-04-28", "movements": [] }, { "institution": "EU AI Office", "status": "Monitoring — no permanent senior appointments; full enforcement from Aug 2, 2026", "last_checked": "2026-04-28", "movements": [] }, { "institution": "Canadian CAISI", "status": "Monitoring", "last_checked": "2026-04-28", "movements": [] } ], "closed_episodes": [], "version_history": [ { "date": "2026-03-30", "change": "Initial entry — no confirmed movements in week 23–29 March 2026", "reason": "Issue 1 research; institutional rebranding noted as tracking complication", "prior_value": null }, { "date": "2026-04-01", "change": "AISI pipeline status updated: all three bodies in simultaneous leadership transition or reduced capacity", "reason": "Issue 2 M15 research — UK interim director, US restructured, EU no new appointments", "prior_value": "No confirmed senior movements; institutional rebranding complicating tracking" }, { "date": "2026-04-10", "change": "Issue 4 — all three principal AI safety bodies still in leadership transition; AISI on interim leadership (Beaumont); NIST AI Safety Inst under review; no permanent directors approaching August 2026 deadline", "reason": "Issue 4 research — structural vacuum continues", "prior_value": null }, { "date": "2026-04-28", "change": "carry-forward-verified + CAISI activity update", "reason": "UK AISI: Adam Beaumont Interim Director confirmed (transformernews.ai Oct 2025; LinkedIn AISI Oct 2025) — no permanent appointment. US CAISI: MOU with GSA March 18 2026 (nist.gov), CRADA with OpenMined March 27 2026 (nist.gov), AI Agent Standards Initiative Feb 17 2026 (nist.gov) — active institution with reduced safety mandate vs Biden-era. EU AI Office: no enforcement decisions yet; full authority from Aug 2 2026.", "prior_value": null } ], "lab_movements": [ {} ], "government_ai_bodies": [ {} ], "revolving_door": [ {} ] }, "baseline": { "uk": { "name": "AI Security Institute (rebranded from AISI, early 2025)", "note": "Adam Beaumont confirmed as Interim Director (October 2025) — ex-GCHQ Chief AI Officer. No permanent director. Mandate shift: safety evaluation → security-focused oversight. Conducted Grok 4 evaluation (CBRN uplift confirmed). Track staff changes — departures/appointments carry higher significance in unstable institutional period.", "last_verified": "2026-04-28", "verification_source": "https://www.transformernews.ai/p/exclusive-uk-aisi-hires-gchq-chief-adam-beaumont" }, "us": { "name": "NIST CAISI (Center for AI Standards and Innovation) — rebranded from AI Safety Institute", "note": "Active in 2026: AI Agent Standards Initiative (Feb 17), MOU with GSA for USAi federal procurement (Mar 18), CRADA with OpenMined (Mar 27). Trump deregulatory pivot reduced enforcement mandate vs Biden era. NIST AI RMF 1.0 in effect but lacks executive enforcement mandate.", "last_verified": "2026-04-28", "verification_sources": [ "https://www.nist.gov/news-events/news/2026/02/announcing-ai-agent-standards-initiative-interoperable-and-secure", "https://www.nist.gov/news-events/news/2026/03/caisi-signs-mou-gsa-boost-ai-evaluation-science-federal-procurement-through" ] }, "eu": { "name": "EU AI Office", "url": "digital-strategy.ec.europa.eu/en/policies/ai-office", "note": "Operational, no binding enforcement decisions yet. Full enforcement authority from August 2, 2026. First supervisory decision = landmark M09+M10 item. GPAI obligations have been enforceable since August 2025; enforcement actions begin August 2026.", "last_verified": "2026-04-28" }, "last_updated": "2026-04-28" } } }, { "name": "Energy Wall", "slug": "energy-wall", "state": "absent", "modules": [ "module_3" ], "absent_message": "No energy-wall signal this cycle. The filter remains armed; the next cycle that emits real content will populate this section. Last cycle checked: issue 38 on 2026-06-05T09:00:00Z.", "fields": null }, { "name": "Capability Watch", "slug": "capability-watch", "state": "Active", "modules": [ "module_12" ], "absent_message": null, "fields": [ "The Scientific Panel creates a technical enforcement gate that labs cannot bypass through legal interpretation. If the Panel classifies a model as GPAI with systemic risk, the lab must comply with Article 55 obligations or exit the EU market. This is a hard enforcement mechanism, not a soft coordination process." ] }, { "name": "Voluntary Commitment Downgrade", "slug": "voluntary-commitment-downgrade", "state": "absent", "modules": [ "module_10", "module_11" ], "absent_message": "No voluntary-commitment-downgrade signal this cycle. The filter remains armed; the next cycle that emits real content will populate this section. Last cycle checked: issue 38 on 2026-06-05T09:00:00Z.", "fields": null } ], "filter_count": 6, "active_count": 4 };

Standards Vacuum EU / China Law & Litigation

Live evidence — persistent_state.module_5_standards_vacuum

Ciyuan Signal EU / China

Live evidence — persistent_state.module_5_ciyuan

AISI-to-Lab Pipeline Personnel

Live evidence — persistent_state.module_15_aisi_pipeline + aisi_status_2026

Energy Wall Investment & M&A

Live evidence — report.module_3.energy_wall

Capability Watch Technical Standards

Live evidence — report.module_12.asymmetric_flags

Voluntary Commitment Downgrade AI Governance Ethics

Live evidence — report.module_11.items

All 16 Module Tags — filter reference