Scientists don't agree yet on approach for recursive self-improvement

Scientists don't agree yet on approach for recursive self-improvement | I spent the last week doing RSI reviews, recursive self-improvement reviews. The scientists do not agree on the exact approach to work yet... I think it's too early to know that question

I spent the last week doing RSI reviews, recursive self-improvement reviews. The scientists do not agree on the exact approach to work yet... I think it's too early to know that question

Beneficiaries (23)

Adverse (6)

{
  "nia": false,
  "url": "https://www.youtube.com/watch?v=DpwmmXmzvfo",
  "mode": "THESIS",
  "role": "Guest-CEO",
  "caveats": "evidence it will work in limited cases",
  "context": "The scientists do not agree on the exact approach to work yet... too early to know",
  "to_year": 2026,
  "verbatim": "I spent the last week doing RSI reviews, recursive self-improvement reviews. The scientists do not agree on the exact approach to work yet... I think it's too early to know that question",
  "direction": "NOT_HAPPEN",
  "from_year": 2026,
  "timeframe": "present",
  "conv_level": "HIGH",
  "milestones": [
    {
      "kind": "llm_pre_event",
      "label": "Karpathy releases open-source AutoResearch RSI loop",
      "notes": "HIT — AutoResearch ran ~700 experiments in 2 days with 11% training speedup; concrete example of working RSI approach.",
      "source": "https://www.thefai.org/posts/on-recursive-self-improvement-part-i",
      "status": "hit",
      "weight": 0.4,
      "ordinal": -7,
      "source_id": null,
      "confidence": 0.9,
      "source_url": "https://www.thefai.org/posts/on-recursive-self-improvement-part-i",
      "expected_date": "2026-03-15",
      "observed_date": "2026-03-15",
      "research_origin": "deep_research",
      "measurement_criterion": "Andrej Karpathy publishes AutoResearch where AI agent autonomously edits PyTorch code, runs trainings, evaluates and loops"
    },
    {
      "kind": "prereq",
      "label": "Nvidia quadrupled chip production output while only doubling human headcount — achieved by deploying AI coding tools (Cursor, Claude Code) a",
      "status": "hit",
      "weight": 0.5,
      "ordinal": -6,
      "source_id": "SEM_012",
      "expected_date": "2026-04-29",
      "observed_date": "2026-04-29"
    },
    {
      "kind": "prereq",
      "label": "Training runs costing $10 billion for a single model will commence sometime in 2025.",
      "status": "hit",
      "weight": 0.5,
      "ordinal": -5,
      "source_id": "SEM_008",
      "expected_date": "2026-04-29",
      "observed_date": "2026-04-29"
    },
    {
      "kind": "prereq",
      "label": "2025 will be the definitive year that agentic systems finally hit the mainstream.",
      "status": "hit",
      "weight": 0.5,
      "ordinal": -4,
      "source_id": "SEM_042",
      "expected_date": "2026-04-29",
      "observed_date": "2026-04-29"
    },
    {
      "kind": "prereq",
      "label": "Recursive self-improvement is already happening now (no longer three years out)",
      "status": "hit",
      "weight": 0.5,
      "ordinal": -3,
      "source_id": "238_009",
      "expected_date": "2026-04-29",
      "observed_date": "2026-04-29"
    },
    {
      "kind": "llm_pre_event",
      "label": "ICLR 2026 RSI workshop convenes with peer-reviewed methods papers",
      "notes": "HIT — workshop scheduled at ICLR 2026 acknowledges scientists still disagree on methods; supports prediction's premise.",
      "source": "https://recursive-workshop.github.io/",
      "status": "hit",
      "weight": 0.4,
      "ordinal": -2,
      "source_id": null,
      "confidence": 0.99,
      "source_url": "https://iclr.cc/virtual/2026/workshop/10000796",
      "expected_date": "2026-04-30",
      "observed_date": "2026-04-25",
      "research_origin": "deep_research",
      "measurement_criterion": "ICLR 2026 Workshop on AI with Recursive Self-Improvement holds session with accepted papers proposing methods"
    },
    {
      "kind": "llm_pre_event",
      "label": "Recursive Superintelligence public launch (raised $500M)",
      "source": "https://ai2.work/blog/recursive-superintelligence-lands-500m-to-build-self-improving-ai",
      "status": "overdue",
      "weight": 0.4,
      "ordinal": -1,
      "source_id": null,
      "confidence": 0.7,
      "source_url": "https://ai2.work/blog/recursive-superintelligence-lands-500m-to-build-self-improving-ai",
      "expected_date": "2026-05-15",
      "miss_emitted_at": "2026-05-30T22:15
... (truncated)