Once 2027 AGI arrives (AI researchers capable of autonomous research), the intelligence explosion begins — compressing roughly a decade of human-led algorithmic progress into a single year and culminating in Superintelligence by 2030.

Once 2027 AGI arrives (AI researchers capable of autonomous research), the intelligence explosion begins — compressing roughly a decade of human-led algorithmic progress into a single year and culminating in Superintelligence by 2030. | First public recursive-self-improvement demonstration

Key catalyst: First public recursive-self-improvement demonstration

Watch events: Frontier lab disclosure of recursive-self-improvement capability; automated ML research benchmarks

Aschenbrenner Situational Awareness framework (2024) continues driving hyperscaler capex and national-security planning; recursive-self-improvement empirical onset is the binary test.

Adverse (4)

{
  "nia": false,
  "qty": "10-year compression to 1 year",
  "mode": "FORECAST",
  "role": "Cited-VC/Researcher",
  "context": "Extends SEM_002 (AGI 2025-2026) and INF_001 (AGI 2027) into the specific ASI-2030 endpoint with explicit decade-in-a-year compression mechanism. Most mathematically rigorous short-timeline framework.",
  "to_year": 2030,
  "conv_cues": "OOMs mathematical framework; explicit year; specific compression ratio",
  "direction": "HAPPEN",
  "from_year": 2027,
  "timeframe": "2027-2030",
  "conv_level": "HIGH",
  "milestones": [
    {
      "kind": "llm_pre_event",
      "label": "Stanford AI Index 2027 documents recursive self-improvement evidence in deployed systems",
      "source": "Stanford AI Index 2026 Technical Performance",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -8,
      "source_id": null,
      "confidence": 0.65,
      "source_url": "https://hai.stanford.edu/ai-index/2026-ai-index-report/technical-performance",
      "expected_date": "2027-04-30",
      "research_origin": "deep_research",
      "measurement_criterion": "Stanford HAI's annual AI Index includes a chapter specifically documenting RSI behaviors in deployed agents, citing >=3 named systems"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q1 window check-in (25%)",
      "status": "pending",
      "weight": 0.05,
      "ordinal": -7,
      "source_id": null,
      "expected_date": "2027-09-17",
      "observed_date": null
    },
    {
      "kind": "llm_pre_event",
      "label": "ICLR / NeurIPS RSI workshop cohort produces published 'AI researcher' agent that authors a top-tier ML paper",
      "source": "ICLR 2026 Workshop on AI with Recursive Self-Improvement",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -6,
      "source_id": null,
      "confidence": 0.6,
      "source_url": "https://recursive-workshop.github.io/",
      "expected_date": "2027-10-01",
      "research_origin": "deep_research",
      "expected_date_range": {
        "to": "2028-06-30",
        "from": "2027-01-01"
      },
      "measurement_criterion": "Peer-reviewed ML paper (NeurIPS/ICLR/ICML) where the primary research/coding contributor is an autonomous agent, validated by program committee"
    },
    {
      "kind": "llm_pre_event",
      "label": "First public recursive-self-improvement demonstration on safety-critical benchmark",
      "source": "METR research program; HyperAgents reference (March 2026)",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -5,
      "source_id": null,
      "confidence": 0.6,
      "source_url": "https://metr.org/",
      "expected_date": "2028-03-16",
      "research_origin": "deep_research",
      "expected_date_range": {
        "to": "2028-12-31",
        "from": "2027-06-01"
      },
      "measurement_criterion": "METR or equivalent independent evaluator publishes report showing measurable capability gains from agent's autonomous self-modification with reproducible methodology"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q2 window check-in (50%)",
      "status": "pending",
      "weight": 0.05,
      "ordinal": -4,
      "source_id": null,
      "expected_date": "2028-06-03",
      "observed_date": null
    },
    {
      "kind": "llm_pre_event",
      "label": "Frontier-lab model card discloses substantive role of automated AI research in capability gain",
      "source": "Aschenbrenner Situational Awareness — intelligence explosion thesis",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -3,
      "source_id": null,
      "confidence": 0.5,
      "expected_date": "2028-06-15",
      "research_origin": "deep_research",
      "expected_date_range": {
        "to": "2029-06-30",
        "from": "2027-06-01"
      },
      "measurement_criterion": "OpenAI / Anthropic / DeepMind model card or system card explicitly attributes >=20% of capability improvement to AI-driven research/algorithmic discovery"
  
... (truncated)