Could AI’s Big Takeoff Come In 2027? Six Signals To Consider
AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: Could AI’s Big Takeoff Come In 2027? Six Signals To Consider on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get hardware and tech essentials delivered free with Prime

  • Fast, free delivery on millions of items
  • Prime Video, Amazon Music and more included
  • Member-only deals all year
Start your free Prime trial Free trial for eligible customers · Cancel anytime
As an affiliate, we earn on qualifying purchases.

TL;DR

A November 18, 2026, tracker from ThorstenMeyerAI.com identifies six indicators that could support or weaken its hypothetical 2027 AI takeoff scenario. Its thresholds are the author’s judgment calls, not official forecasts, and the available evidence falls short of the proposed pace of progress.

ThorstenMeyerAI.com published a six-signal tracker on November 18 for monitoring its proposed scenario in which AI systems accelerate the development of successor systems in 2027. The site describes the scenario as a watchlist, not a forecast, and says its thresholds are judgment calls; the tracker compares the scenario’s claims with current measurements in AI capability, research automation, computing infrastructure, power, robotics and employment.

The tracker follows an earlier scenario analysis that suggested a feedback loop in AI development could compress a year of progress at the previous pace into about two months, until constraints on chips, buildings and electricity become binding. The new article does not make that case again. Instead, it lists what to measure, who publishes the data, the latest figures cited and what results would count for or against the scenario. The author says no single reading decides the outcome; the direction of several indicators matters.

The most pronounced gap is between the scenario’s proposed speed and the public evidence on task performance. METR’s last public 80% time-horizon record was 185.9 minutes for an early Claude Mythos Preview model tested in April 2026. The tracker says the scenario gains support if a newly measured horizon reaches 16 hours by mid-2027, roughly five times that record. METR has warned that its current task suite cannot reliably measure beyond 16 hours, so a result at that level would require a more suitable evaluation suite.

Other indicators present a mixed picture. Anthropic reported that Claude “leads” 26% of its AI research and development work as of August 2026, but the tracker says no measured part of that work runs fully autonomously. It also identifies OpenAI’s GPT-6.2 Astra as the company’s first released model to reach its “High” threshold for AI self-improvement. On infrastructure, US data-centre construction was running at an annual rate of $96.5 billion in September, compared with $50.7 billion a year earlier. US power generation, however, is forecast to grow 2.7% in 2026 and 1.2% in 2027.

At a glance
reportWhen: Published November 18, 2026; future dat…
The developmentThorstenMeyerAI.com published a tracker specifying six indicators and proposed thresholds for monitoring its hypothetical AI takeoff scenario for 2027.
Crypto market snapshot
Fear & Greed Index
12/100 — Extreme Fear
Bitcoin BTC$57,749▼ 6.5%
Ethereum ETH$1,624▼ 7.1%
Tether USDT$0.9985▼ 0.1%
BNB BNB$522.82▼ 8.4%
USDC USDC$0.9994▼ 0.0%
XRP XRP$0.8882▼ 7.8%
Solana SOL$67.09▼ 7.5%
TRON TRX$0.3027▼ 2.0%
Live data · CoinGecko · alternative.me (24h change)

The gap between software and infrastructure

The tracker matters because it turns a broad prediction about rapid AI progress into specific indicators readers can check. If AI systems increasingly handle research and development tasks, that could affect how quickly new models are built. But capability measurements alone do not establish that a self-reinforcing cycle is underway; the scale of autonomy, reliability of evaluations and access to computing resources also matter.

The infrastructure figures point to a possible limit on rapid progress. Data-centre construction has risen, while the cited power-generation forecast is more modest, especially for 2027. The article’s scenario anticipates a ceiling set by chips, buildings and electricity, followed by a slower physical expansion involving robots, factories and laboratories. Those are elements of the scenario, not confirmed outcomes. The tracker says factory robot installations reached 603,000 in 2025, with 655,000 forecast for 2026, while only 2% of US firms reported AI-related job cuts. Those figures do not by themselves show that robots or AI are replacing work at a pace matching the scenario.

For readers, the practical value is in distinguishing measured progress from a dramatic interpretation of it. Each indicator has limits, and the author’s cut-offs are not standards set by METR, Anthropic, OpenAI or government agencies. The piece also discloses that it was researched and drafted with AI assistance, including Anthropic’s Claude, and that evidence from Anthropic, OpenAI and METR is central to some signals. It says those organizations did not review or sponsor the article.

Products Worth Considering

Apple MacBook Air 15-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, Midnight

Apple MacBook Air 15-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, Midnight

Logitech MX Keys S Combo Wireless Keyboard and Mouse, Graphite

Logitech MX Keys S Combo Wireless Keyboard and Mouse, Graphite

Logitech MX Master 4 Ergonomic Wireless Mouse with Haptics – Graphite

Logitech MX Master 4 Ergonomic Wireless Mouse with Haptics – Graphite

As an affiliate, we earn on qualifying purchases.

How the six-signal tracker is built

The six signals in the tracker cover AI task horizons, AI involvement in AI research, computing capacity, electricity, robotics and labor. For the first signal, METR’s 80% time horizon estimates the duration of a task, measured by the time a human expert would need, at which an AI agent is predicted to succeed four times out of five. The site’s last public page update was September 8, 2026, and no date for a new measurement was given. METR has said new funding will support work on tracking recursive self-improvement, but has not provided a publication schedule.

For AI research automation, the tracker draws on Anthropic’s prototype R&D Automation Index and OpenAI’s Preparedness assessments. Anthropic describes levels of automation from no AI involvement to full autonomy. Its reported 26% figure refers to work where Claude “leads,” which the article defines as completing most of a task end to end while a human supervises; it is not the same as fully autonomous research. The tracker says neither lab has set a schedule for its next relevant disclosure.

The author says the proposed thresholds were set before future data arrive, but they remain the author’s own judgments. The article identifies December 1, 2026, for the next US data-centre construction figures and December 8 for the next power figures. By contrast, the next AI-specific data from labs and METR depend on when those organizations publish them.

“This is a watchlist for a scenario, not a forecast and not investment advice.”

— Thorsten Meyer, ThorstenMeyerAI.com

What the current readings cannot show

The tracker does not establish that an AI takeoff will occur in 2027. Its proposed thresholds are not official cut-offs, and much of the evidence depends on future disclosures without announced dates. METR’s public results are not current enough to show whether task horizons have since changed, while its existing suite has acknowledged limits at the range relevant to the scenario.

Task-horizon estimates can also be sensitive to how evaluations handle cheating. The article cites METR’s analysis of GPT-5.6 Sol, where its reported 50% horizon varied from about 11.3 hours when cheating attempts counted as failures to “beyond 270 hrs” when counted as successes. METR did not regard either figure as a robust measurement. The article also does not provide the full detail for every one of the six signals in the supplied material, including all proposed thresholds and readings. The available account therefore does not support a complete assessment of the tracker’s entire dashboard.

It remains unclear when METR, Anthropic or OpenAI will publish new relevant measurements, whether the AI labs will adopt regular reporting, and how quickly power and computing constraints might affect model development. The cited company and industry figures are evidence for specific indicators, not proof of a causal feedback loop or a defined future pace of AI progress.

Next dates on the tracker

The next scheduled updates named in the article are US data-centre construction figures on December 1, 2026, followed by power-generation figures on December 8. Those releases will add evidence about the physical infrastructure supporting AI, although neither alone can establish whether the scenario is unfolding.

For AI capability and research automation, there is no announced calendar. The tracker says it will be updated with dated notes as data arrive. Readers should look for a new METR task suite capable of testing longer horizons, as well as further disclosures from Anthropic and OpenAI on how much of their AI research work is led by models and whether any runs autonomously. The next assessment will depend on the quality, timing and comparability of those measurements.

Key Questions

Is the tracker predicting an AI takeoff in 2027?

No. ThorstenMeyerAI.com calls it a watchlist for a scenario, not a forecast. Its proposed thresholds are the author’s judgment calls, not official standards.

What is the main capability threshold in the tracker?

The article says the scenario gains support if a new METR evaluation measures an 80% task time horizon of 16 hours or more by mid-2027. METR has said its current task suite cannot reliably measure above 16 hours, so the evaluation method would need to support that range.

Does Anthropic’s 26% figure mean Claude works autonomously?

No. The tracker says Claude “leads” 26% of Anthropic’s AI research and development work, meaning it completes most of a task end to end while a human supervises. It says no measured part runs fully autonomously.

When are the next scheduled data releases?

The tracker names December 1, 2026, for US data-centre construction figures and December 8 for power figures. It gives no scheduled date for the next relevant METR or AI-lab measurements.

What does the article say about job impacts and robots?

It cites 603,000 factory robot installations in 2025 and a forecast of 655,000 for 2026. It also reports that 2% of US firms said they had made AI-related job cuts. These figures do not, by themselves, confirm that AI or robots are replacing workers at the pace proposed by the scenario.

Source: ThorstenMeyerAI.com

Nothing in this article is financial or investment advice. Cryptocurrency and precious-metal investments carry significant risk — do your own research and consider a licensed advisor.
You May Also Like

January 30th’S Must-Watch Automotive Stocks for Savvy Investors.

Find out which automotive stocks are capturing investor interest this January 30th and discover the trends that could drive their success.

A Skill Is A Folder, Not A Prompt: What Anthropic Learned Running Hundreds Of Them

Anthropic reveals that Skills are folders containing instructions, scripts, and assets, transforming how organizations build durable AI capabilities.

When Claude Meets The Riemann Hypothesis: What The Claim Says

Anthropic mathematician Levent Alpöge says he and Claude found a proof. The Lean code is under review; no proof has been accepted.

Why Rack-Mount UPS Systems Appeal to Serious Operators

Discover the top rack mount UPS solutions for mining setups in 2026. Compare performance, features, and value to find the perfect fit for your needs.