🔍 Read the full analysis: Checking The Signs Of Superintelligence: 50 Clues About AI And AGI on ThorstenMeyerAI.com
Get smart everyday buys 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
AI consultant Alan D. Thompson’s public tracker marks two of 50 proposed signs of artificial superintelligence as achieved, 33 as partially achieved and 15 as not achieved. A review of 18 prominent claims found some independently supported advances, especially in mathematics, but other claims rely on company statements or broad definitions of progress. The tracker is one researcher’s framework, not a measured consensus that superintelligence has arrived.
A review published by ThorstenMeyerAI.com on October 26, 2026, checked 18 prominent claims in AI consultant Alan D. Thompson’s list of 50 signs of artificial superintelligence against research papers, lab statements and other primary sources. Thompson’s tracker marks two signs achieved, 33 partially achieved and 15 not achieved; the review says the checked claims include independently supported mathematical results, vendor-reported milestones and disputed or weakly sourced assertions. The tracker is an individual framework, not evidence of a consensus that superintelligence has arrived.
Thompson, an Australian AI consultant who publishes The Memo newsletter, maintains a public checklist covering developments from recursive code optimization to much broader scientific and technological goals. His October video, “ASI: A better world – 2026,” surveys entries from January through October. In his tracker, he describes Claude Mythos Preview and GPT-6 Astra as “early ASI” systems. That is Thompson’s assessment, not a designation established by independent agreement.
The review’s author says the tracker’s orange “partially achieved” category has a broad threshold. In an August correction note, Thompson describes it as including documented material progress such as funded programs, research infrastructure, formal scientific missions and enabling work. That means a partial rating does not necessarily indicate that the stated outcome is close. The review gives the example of two exoplanet candidates reportedly found with Claude Code being placed under a tracker indicator about colonizing a planet beyond Earth.
Of the 18 claims examined, the review classifies seven as verified, five as vendor claims, three as based on a single source and three as disputed or contradicted. It says the strongest independently checkable evidence clusters around mathematical work, a validated physics calculation and regulatory filings. In mathematics, it cites contributions described by Terence Tao, Don Knuth and Tim Gowers, while retaining caveats about human coaching, relevant literature and the kinds of problems involved. The supplied review material does not provide the full evidence or verdict for every claim in its selection.
Whose superintelligence?
One word, very different bars
Dr Alan D. Thompson's ASI starts where Google DeepMind's scale puts "Expert AGI". The same model can be "early ASI" on one scale and a strong but narrow AGI candidate on another.
- Median human50th percentileThompson: AGIa machine at the level of an average (median) humanDeepMind: Competent AGIat least the 50th percentile of skilled adults
- Expert90th percentileThompson: ASIexpert-human level in practically any fieldDeepMind: Expert AGIat least the 90th percentile of skilled adults
- Virtuoso99th percentile—DeepMind: Virtuoso AGIat least the 99th percentile of skilled adults
- Beyond every human100%—DeepMind: Superhuman (Level 5)outperforms 100% of humans
- Beyond organisations2026 report—DeepMind: ASImore capable than large organisations of humans
Thompson’s tracker status as of 26 October 2026. DeepMind rows: “Levels of AGI” (2023) and, for the last row, “From AGI to ASI” (June 2026). OpenAI’s Charter: AGI = “highly autonomous systems that outperform humans at most economically valuable work”.
Sources: LifeArchitect.ai ASI checklist (26 Oct 2026), AGI countdown, Morris et al., Levels of AGI (2023), Genewein et al., From AGI to ASI (2026), OpenAI Charter.
Powered by Thorsten Meyer AI
What the Checklist Can—and Cannot—Show
The tracker draws attention to developments worth monitoring, but its totals depend on how each sign is defined and what counts as progress. Because partial status can include enabling work, 33 partial indicators do not mean 33 near-complete breakthroughs. Readers should distinguish a research program moving forward from the outcome named by an indicator actually being achieved.
The review also shows why source quality matters. A result described in a company post may be a meaningful lead, but it is not the same as a claim independently reproduced or checked by outside experts. By separating verified results from vendor claims, single-source reports and disputed claims, the review offers a more cautious reading than treating every item in a fast-moving montage as established fact.
One benchmark mentioned in the source material concerns AI-assisted research at Anthropic. The company says its best internal model scores 62.8% on a test where it estimates at least 85% would be needed to replace its researchers fully. Anthropic also says its research and development is faster with AI, but “not yet by a factor of 2.” Those are company-reported assessments, not independent proof that AI can autonomously improve itself at a transformative rate.
Products Worth Considering
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
Apple iPad Air 13-inch (M4), 128GB, Wi-Fi 7, Space Gray
Canon imagePROGRAF PRO-1100 Professional 17-Inch Wireless Inkjet Photo Printer Bundle
As an affiliate, we earn on qualifying purchases.
Different Definitions, Different Thresholds
Thompson defines AGI as a machine performing at the level of an average, or median, human, and ASI as a machine performing at expert-human level in practically any field. Those definitions affect what his labels imply. Google DeepMind’s 2023 “Levels of AGI” framework separates “Competent,” “Expert,” “Virtuoso” and “Superhuman” performance, with Expert defined as around the 90th percentile of skilled adults. Under that framework, Thompson’s ASI definition is closer to Expert AGI than to superhuman performance.
Google DeepMind’s June 2026 report, “From AGI to ASI,” uses a more demanding description of superintelligence: systems more intelligent and cognitively capable than large organisations of humans. OpenAI’s Charter defines AGI around highly autonomous systems that outperform humans at most economically valuable work. These approaches emphasize different things: expert performance in fields, capability beyond groups of people, or autonomy across economically important work.
Thompson also runs a separate countdown to AGI that he puts at 99%. He says the remaining percentage depends on a robot completing multi-step fine assembly, such as putting together flat-pack furniture, with average-human reliability. That figure is his own estimate and depends on his chosen threshold; it is not a standardized measurement used across AI researchers.
““documented material progress toward an indicator, including funded programs, research infrastructure, formal scientific missions, and enabling work””
— Alan D. Thompson, in an August correction note about the tracker
How Much Progress Is Independently Checked?
The review’s verdicts apply to the 18 selected claims, not necessarily all 50 indicators. The source material does not give enough detail to independently assess every entry or reproduce the review’s complete claim-by-claim evidence. Its categories also combine different kinds of evidence, from formal results to company-reported metrics, so they should not be treated as a single standardized test of superintelligence.
The review says some tracker entries have not materialized in observable ways: it identifies no household robots at scale, no verifiably billion-dollar AI-run company and no visible break in the U.S. unemployment rate. Those observations do not by themselves settle whether AI capabilities meet any chosen definition of AGI or ASI. They point instead to limits in the evidence for broad, real-world autonomy and economic effects.
It remains unclear whether the systems Thompson calls “early ASI” meet a threshold that other researchers would accept, how robust their reported achievements are outside selected tasks, and how much human guidance was involved in each result. The review also notes that some mathematics examples involved coaching, literature searches or problem formulations that required clarification. Those qualifications matter when interpreting claims of autonomous discovery.
Evidence to Watch in Future Updates
The next useful test will be whether new claims are accompanied by materials that outside researchers can inspect: papers, reproducible methods, formal proof certificates, validated measurements or official filings. For claims about practical autonomy, readers will also need evidence that systems complete extended tasks reliably without substantial human intervention, rather than only performing well on a narrow benchmark.
Thompson’s tracker is dated October 26, 2026, and can be compared with later updates to see which indicators move from partial to achieved and what evidence supports each change. The source material does not identify a scheduled next update or a formal independent review of the checklist. Until more is published, the reported status remains a snapshot of one researcher’s assessment, while the broader question of whether AI has reached AGI or superintelligence remains contested.
Key Questions
Has artificial superintelligence been confirmed?
No consensus confirmation is reported. Thompson calls some current models “early ASI” under his own definition, but the review presents that as his assessment and notes that definitions vary among researchers and labs.
What does Thompson’s tracker say has been achieved?
As of October 26, 2026, the tracker marks two of 50 indicators achieved, 33 partially achieved and 15 not achieved. Thompson’s partial category includes enabling work and funded programs, not only outcomes close to completion.
How many claims did the review verify?
The review examined 18 selected claims: it classified seven as verified, five as vendor claims, three as based on a single source and three as disputed or contradicted. These figures describe that selection, not all entries in the 50-sign tracker.
Why do definitions of ASI matter?
Different definitions set different bars. Thompson describes ASI as expert-human performance in practically any field, while Google DeepMind’s 2026 report describes systems more capable than large organisations of humans. A system may fit one definition without meeting another.
Source: ThorstenMeyerAI.com
