🔍 Read the full analysis: Will AI Take Off Next Year? Six Signals To Keep An Eye On on ThorstenMeyerAI.com
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
ThorstenMeyerAI.com published a watchlist on Nov. 18, 2026, to track a scenario in which AI helps build more capable AI systems and speeds progress in 2027. Its six signals cover AI task performance, lab automation, compute, power, robotics and employment; the author stresses that the thresholds are judgment calls, not a forecast, and several readings have no scheduled release date.
ThorstenMeyerAI.com published a six-signal tracker on Nov. 18 for testing its scenario that AI systems could help build better AI systems quickly enough to accelerate progress in 2027. The article is a watchlist, not a forecast: it sets the author’s own thresholds for evidence, while noting that some key measures are infrequently updated or cannot yet be measured reliably.
The tracker follows AI task time horizons, automation of AI research, compute, electricity, industrial robots and employment. Its author says no single reading would settle the scenario; the case would grow stronger if several indicators moved in the expected direction. The thresholds are described as judgment calls rather than official standards, and for compute and labor the article says no publisher defines a threshold.
The clearest gap is between the scenario’s proposed pace and public task-performance data. The article cites METR’s last public 80% time-horizon record, 185.9 minutes for an early Claude Mythos Preview model in April 2026. Its scenario threshold is 16 hours by mid-2027, roughly five times that record. METR says its current task suite cannot measure reliably above 16 hours, and its public results page is no longer actively updated.
For AI research automation, the article reports Anthropic’s August 2026 estimate that Claude “leads” 26% of the company’s AI research and development work, where “leads” means completing most of a task end to end under human supervision. It also says no measured portion runs fully autonomously. OpenAI’s GPT-6.2 Astra is described as its first released model to reach the company’s “High” threshold for AI self-improvement. These are lab-reported assessments, not independent demonstrations that AI is building future systems without human involvement.
Testing the Pace of AI Progress
The tracker matters because it turns a broad claim about accelerating AI into specific evidence readers can monitor. If task horizons lengthen, AI takes on more of research work, and infrastructure expands in parallel, the scenario’s feedback-loop argument would gain support. If those indicators remain limited, the proposed pace would be harder to substantiate.
The infrastructure signals also point to a potential constraint. The source reports US data-centre construction at an annual rate of $96.5 billion in September 2026, compared with $50.7 billion a year earlier. It cites expected US power-generation growth of 2.7% in 2026 and 1.2% in 2027. Those figures suggest construction and electricity supply may move at different speeds, but they do not by themselves establish how much computing capacity will become available or when.
Robotics and employment provide a longer-term check on whether AI advances are spreading into the physical economy. The article cites 603,000 industrial robot installations worldwide in 2025 and a forecast of 655,000 for 2026, alongside a report that 2% of US firms had AI-related job cuts. These measures have different scopes and do not prove that AI caused broader changes in work. Together, they help distinguish rapid software progress from slower changes in factories and employment.
Products Worth Considering

LG 27GS60QC-B Ultragear 27-inch Curved Gaming Monitor (QHD, 180Hz, 1ms, 1000R)

Samsung 32 Odyssey G5 QHD Gaming Monitor

Bose QuietComfort Headphones (2nd Gen), Wireless Noise-Cancelling Over-Ear Headphones, Black
As an affiliate, we earn on qualifying purchases.
How the Six Signals Are Set
The tracker follows a separate scenario article published earlier by ThorstenMeyerAI.com. That scenario proposes that AI systems may increasingly perform work involved in developing the next generation of AI, with progress potentially constrained by chips, buildings and power before robotics and factories become more important. The Nov. 18 article says it does not repeat the full argument; it records what to measure and what evidence would shift the scenario’s standing.
The author explicitly labels the tracker opinion and scenario analysis, not investment advice. The author also discloses that AI tools, including Anthropic’s Claude, assisted with research and drafting, and that evidence from Anthropic, OpenAI and METR features in the analysis. The source says none of those organizations reviewed or sponsored the article.
Only two of the six signal groups have announced dates for near-term figures: US construction data on Dec. 1 and power data on Dec. 8, 2026. The source says lab assessments and METR measurements will appear when their publishers choose to release them. That uneven schedule limits how quickly the scenario can be tested.
“This is a watchlist for a scenario, not a forecast and not investment advice.”
— ThorstenMeyerAI.com, describing the tracker
Where the Evidence Falls Short
The source does not provide the complete details for all six signals in the material supplied. It identifies the topics and gives readings for task horizons, lab automation, infrastructure, power, robotics and employment, but the final portion of its description of Anthropic’s automation scale is cut off. The precise thresholds and measurement methods for every signal therefore cannot be fully assessed from the supplied text.
There is also no public timetable for the next METR time-horizon result or for new AI-lab automation and self-improvement assessments. METR’s page is described as inactive, and its present task suite has limits at the range relevant to the scenario. Lab measures are also based on companies’ own assessments. Independent comparability and verification remain open questions.
The article’s infrastructure and workforce figures are observations or forecasts, not proof that AI progress is accelerating. The source does not establish that new data-centre construction will translate directly into usable computing power, or that the reported job cuts were caused solely by AI. The scenario’s outcome, the timing of further releases and how the author will revise the tracker remain uncertain.
The December Data Checkpoints
The next scheduled releases identified by the source are US data-centre construction figures on Dec. 1, 2026, and power-generation figures on Dec. 8. Those updates can show whether the infrastructure indicators are continuing in the directions described, but they will not directly answer whether AI systems are taking over more AI research work.
For that, readers will have to wait for new disclosures from Anthropic and OpenAI, and for any updated METR measurements or task suites. The article says it will add dated notes as data arrive. The key test will be whether multiple indicators move together, not whether one headline number crosses a threshold on its own.
Key Questions
Is the article predicting that AI will take off in 2027?
No. ThorstenMeyerAI.com describes the piece as a watchlist for a scenario, not a forecast. Its thresholds are the author’s judgments about what evidence would strengthen or weaken that scenario.
What are the six signals?
The tracker covers AI task time horizons, AI involvement in research and development, compute, electricity, industrial robots and employment. The source says no single reading decides the scenario.
What is the current public task-horizon record cited?
The article cites METR’s 185.9-minute 80% time horizon for an early Claude Mythos Preview model in April 2026. It sets 16 hours by mid-2027 as a threshold that would support the scenario, while noting METR’s current suite cannot reliably measure beyond that range.
When will the next figures arrive?
The source lists US data-centre construction figures for Dec. 1, 2026, and power-generation figures for Dec. 8. It gives no scheduled date for new METR or AI-lab measurements.
Does Anthropic’s 26% figure mean AI research is autonomous?
No. The source says “leads” means Claude completes most of a task end to end while a human supervises. It also says no measured part of Anthropic’s AI research and development work runs fully autonomously.
Source: ThorstenMeyerAI.com
