📊 Full opportunity report: The Co-Founder’s Black Hole — A Structural Read on Jack Clark’s Automated AI R&D Essay on ThorstenMeyerAI.com — validation score, market gap, and execution plan.
TL;DR
Jack Clark, co-founder of Anthropic, forecasts a >60% probability of achieving autonomous AI research systems by 2028. This prediction highlights significant risks and the current institutional gap in managing such a transition.
On May 4, 2026, Jack Clark, co-founder and head of policy at Anthropic, publicly stated there is a more than 60% probability that AI systems capable of autonomously building their own successors will emerge by the end of 2028. This is the first time a sitting AI lab leader has assigned a specific probability and timeframe to the advent of fully autonomous AI research, marking a significant shift in institutional forecasting and risk assessment.
Clark’s forecast was published in his essay ‘Import AI #455,’ where he synthesizes evidence from multiple benchmarks and technical analyses to argue that the convergence of technological progress and economic incentives makes such autonomous systems likely within the next 32 months. You can read more in Jack Clark Says It Out Loud — Reading the Co-Founder’s 60%/2028 Estimate on Automated AI R&D. He highlights that current AI capability benchmarks are rapidly saturating, with improvements following exponential trajectories that align with reaching the threshold for autonomous research by 2028.
He emphasizes that this forecast is not speculative but based on observable trends in AI performance metrics, including benchmark saturation patterns and compute speedups. Clark also discusses the structural implications, comparing the transition to a ‘black hole,’ where the future beyond the threshold becomes unpredictable and potentially uncontrollable, due to the degradation of forecastability once a certain technological boundary is crossed.
Clark’s statement carries institutional weight, as it is the first public, probabilistic forecast from a prominent AI leader with a clear deadline, effectively committing Anthropic to a specific timeline and risk posture. This has immediate implications for AI policy, funding, and safety research, given the potential for rapid, autonomous AI development within a short window.
The black hole
is visible.
Four threads converge. One window. Anthropic’s head of policy has publicly committed to crossing a civilizational threshold within 32 months.
The structural feature of Clark’s argument is not that we cross a boundary and continue forward; it is that beyond a certain threshold, the forecastability of subsequent events degrades dramatically. We can see the geometry around the threshold. We can estimate when we will reach it. We cannot model what happens on the other side. The black hole event horizon analogy is precise.
Four pieces. One argument.
The four prior pieces in this series each addressed a single thread of Clark’s argument. The threads are independently significant. What this synthesis argues: they converge on a structural finding larger than any individual thread.
AI Tools for Teachers: The Practical Guide to Using Artificial Intelligence to Save Time, Boost Engagement, and Personalize Learning (AI-Productivity Series)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Four threads. Four convergence arguments.
The threads converge structurally rather than independently. Each pair of threads produces a specific structural argument. The aggregate is larger than the parts.
Building Agentic AI Systems: Create intelligent, autonomous AI agents that can reason, plan, and adapt
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Clark’s essay doesn’t say.
Each sub-piece identified per-thread omissions. The synthesis level has its own omissions — features of the integrated argument that don’t appear in any single sub-piece but emerge when the threads are read together. Each is a real coordination problem with no resolution at scale.
Multi-Agent Systems Engineering: Design architecture with evidence: metrics, risk gating, failure modes, and tested reference code—benchmarks, debugging, and production hardening for AI agents
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Thirty-two months. Five markers.
From May 4, 2026 to December 31, 2028 is 32 months. The trajectory either delivers the threshold Clark forecasts or it doesn’t. Specific indicators along the way that resolve the synthesis read in either direction.
- Clark publishes 60%/2028
- METR ~12 hr
- SWE-Bench 93.9%
- CORE solved
- Anthropic IPO prep
- METR ~100hr target
- SWE saturated
- MLE-Bench saturating
- PostTrain 40-50%
- Anthropic IPO Q4
- METR 300-500hr
- MLE saturated
- PostTrain at human
- RSI demo non-frontier
- 30%/2027 evidence
- METR 1K-3K hr
- “Trains successor” demos
- Alignment claims
- Catastrophic-risk window
- Stage 2 visible
- METR ~10K hr (naive)
- Automated AI R&D OR
- Inflection visible
- Machine economy Stage 3
- Black hole crossed
If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Five errors. Honest probabilities.
A serious analysis owes the reader an explicit account of where it could be wrong. Five categories of potential error in the synthesis above. The structural finding survives at lower forecast probabilities but is less acute.
Three parts. One window.
The four threads converge. The synthesis-level omissions sharpen the picture. The structural finding is the answer to “what does the Clark essay actually tell us, and what does it imply we should do?”
The black hole is visible. The event horizon is 32 months out. We can see the geometry around the singularity. We cannot see past it. What we can do during the window is build the institutional response that will determine what we encounter on the other side.
Implications of a Structural Shift in AI Development
This forecast indicates a potential paradigm shift in AI development, where autonomous systems could operate with minimal human oversight, potentially impacting safety, regulation, and research priorities. Understanding these developments is crucial for stakeholders monitoring AI progress. The predicted timeline suggests that the next 32 months may be critical for establishing effective governance and safety protocols, although current institutional capacities may need to be strengthened to address emerging risks.
The analogy of a ‘black hole’ illustrates the challenge: once a certain technological threshold is crossed, predicting future developments becomes increasingly difficult, raising questions about oversight and control of advanced AI systems. Addressing these issues requires proactive policy and safety measures.
Converging Trends in AI Benchmarks and Capabilities
Multiple independent benchmarks over the past two years have shown rapid, exponential improvements in AI capabilities across diverse domains, including research automation, language understanding, and compute efficiency. Notably, the METR time horizon benchmark, which measures the duration of AI tasks, has increased from 30 seconds in 2022 to 12 hours in 2026, indicating progress toward more autonomous research cycles by 2028.
These patterns are supported by advancements in AI training speedups, with systems surpassing human performance in key tasks. The convergence of these trends suggests that the technical feasibility of autonomous AI research is approaching a critical point, consistent with Clark’s forecast and the ‘black hole’ analogy.
While previous forecasts from researchers and industry leaders have been less specific, Clark’s institutional statement reflects a move toward more concrete, probability-based predictions grounded in current data.
“There’s a likely chance (60%+) that no-human-involved AI R&D — an AI system capable of autonomously designing and improving its own successor — occurs by the end of 2028.”
— Jack Clark
Uncertainties Surrounding the Autonomous AI Threshold
While Clark’s forecast is based on current observable data, there remain uncertainties regarding the precise point at which AI systems will achieve full autonomy in research and the potential for unforeseen technical or economic obstacles. The ‘black hole’ analogy underscores that beyond a certain threshold, future developments may become difficult to predict or control.
Additionally, the development of appropriate political, regulatory, and safety frameworks to manage this transition is still underway, which could influence the actual timeline and impact of autonomous AI systems.
Next Steps in Monitoring and Preparing for Autonomous AI
Stakeholders including researchers, policymakers, and industry leaders should continue to monitor advancements in AI benchmarks and computational capabilities. For more insights, see Jack Clark’s forecast on autonomous AI research. Efforts should focus on developing safety protocols, regulatory frameworks, and international cooperation mechanisms to address potential risks associated with rapid autonomous AI development.
Further research is needed to understand the limits of current AI systems and to identify early indicators of approaching the autonomous research threshold. Transparency from leading AI labs will be important for informing policy and safety measures.
Key Questions
What does ‘no-human-involved AI R&D’ mean?
It refers to AI systems capable of independently designing, improving, and deploying new AI models without human intervention, potentially enabling self-sustaining development cycles.
Why is the 2028 deadline significant?
Clark’s forecast suggests that within the next 32 months, AI systems may reach a level where they can conduct research autonomously, raising important considerations for safety and governance.
What are the risks of reaching this autonomous threshold?
Potential risks include reduced human oversight, unpredictable system behavior, and rapid surpassing of human capabilities, which could pose challenges for safety and control.
How reliable is Clark’s forecast?
It is based on current data and trends, but inherent uncertainties in complex systems mean that the exact timing and nature of the transition remain uncertain.
Source: ThorstenMeyerAI.com
