iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Yes. Artificial superintelligence (ASI) might never be achieved, and there is no reliable consensus probability or timeline for it. Recent AI advances make major change plausible, but they do not prove that AI will reach superintelligence—or that any such change will arrive as one sudden transformation. Technical limits, the shape of future research, and choices by governments and societies all matter.
What would count as ASI that radically changes society?
ASI is not a standardized, measurable finish line. Google DeepMind’s 2026 report describes it intuitively as an AI system more intelligent and cognitively capable than large organizations of humans. That framing is useful for discussion, but it does not identify a single benchmark, capability, or date that would settle whether ASI had arrived.
“Radically changes society” is just as open to interpretation. It might mean a sharp break in how people work and govern, or a succession of changes as AI helps produce advances in science and technology. The distinction matters: gradual but far-reaching effects would still be transformative, even if there were no clear day when society crossed into an “ASI era.”
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteIt also helps to distinguish ASI from related claims. A system performing well on a particular task, automating a large share of tasks, or meeting a proposed definition of AGI does not by itself show that it is superintelligent. Predictions about one of those milestones should not be treated as predictions about all the others.
#1 Best Overall
How could AI progress reach ASI—or fail to?
DeepMind’s report discusses four possible pathways from AGI toward ASI. They are scenarios for how progress might happen, not forecasts that it will.
Scaling AGI
One route is to continue expanding the capabilities of AGI systems. That route depends on continued progress, but the report identifies possible frictions and bottlenecks along the way. Their eventual importance is unresolved; it would be premature to assume either that scaling will stop or that it can continue without meaningful obstacles.
Shifts in AI paradigms
Progress could also come from changes in the underlying approaches used to build AI, rather than simply extending current methods. The possibility of a new paradigm leaves room for breakthroughs, but does not tell us whether one will occur, when it might happen, or how much it would improve general capability.
Recommended Free Tools
Rank #2
Recursive improvement
Another proposed path is that AI systems contribute to improving AI research and development, potentially accelerating further progress. This is a possible mechanism, not evidence that a self-reinforcing cycle will begin or remain effective. Whether such improvement could overcome practical frictions is among the open questions identified by DeepMind.
Large-scale multi-agent collectives
A further possibility is combining many AI agents into a coordinated system. The resulting capabilities would depend on how well the agents work together and on the limits that coordination encounters. The existence of this pathway does not establish that a collective would outperform a single system enough to reach ASI.
DeepMind concludes that determining whether frictions along these pathways have negligible or substantial effects raises concrete open research questions. That uncertainty is central: several plausible routes do not add up to a guarantee that any route will succeed.
What does recent AI progress actually tell us?
The International AI Safety Report 2026 describes rapid improvement in some evaluated capabilities, while cautioning that benchmark results are not direct forecasts of broad real-world competence. Performance tends to decline as tasks become longer, and benchmark exercises may leave out complications such as limited resources or incomplete information. A strong result in a defined test is evidence about that test, not proof that a system can reliably handle open-ended work.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →The UK AI Security Institute reports that a model completed some expert-level cyber tasks in 2025 and that performance improved in its autonomy evaluations. It also reports that, within the cyber evaluation tasks it assessed, the duration of tasks models could complete with a 50% chance doubled roughly every eight months. That is a trend in the Institute’s evaluation suite, not a universal law of AI progress. These bounded results indicate meaningful capability gains; they do not establish that ASI exists or must follow.
There is also uncertainty about the overall pace of general-purpose AI progress. The International AI Safety Report interim report records disagreement among experts and notes limitations in technical evaluation and risk reduction. DeepMind cautions that the familiar picture of one abrupt transformation at AGI may be wrong: change could instead unfold through a series of advances.
Do expert surveys tell us how likely ASI is?
Not directly. Surveys can show how experts think about particular questions, but estimates depend on what respondents are asked, who answers, and the assumptions they make. AI Impacts lists 2,778 respondents for its 2023 Expert Survey on Progress in AI. That number describes the survey’s respondent pool; it is not a probability that ASI will happen.
An earlier survey of attendees at three AI conferences, published as Forecasting Transformative AI: An Expert Survey in 2019, reported median forecasts that AI could automate 90% of current human tasks with a 50% probability within 25 years, and with a 99% probability within 50 years. Those are respondents’ forecasts about task automation—not estimates of the chance of ASI. The survey also found that conference attendance significantly affected forecasts, a reminder that the sample and method shape results.
To assess any forecast, first check what outcome it predicts: AGI, task automation, transformative AI, or ASI. Then consider its time horizon, respondent sample and survey method, assumptions about scaling and algorithmic progress, and whether it addresses technical capability, deployment, or societal impact. Without those distinctions, numbers that look comparable may be answering different questions.
Best Value
Why is the outcome not inevitable?
The UK government’s International Scientific Report interim report puts the point plainly: “nothing about the future of AI is inevitable.” Its work represents 75 contributing AI experts and an advisory panel nominated by 30 countries, the EU, and the UN. The report describes uncertainty about progress and emphasizes that social and government choices help shape AI development and risk.
Those choices do not settle the technical question of whether ASI can be built. They do affect the path AI takes: what development is pursued, how risks are assessed and reduced, and how systems are deployed. A society-changing future is therefore not just a question of what capabilities might be technically possible; it also depends on how people and institutions respond.
What can be concluded about the chance of ASI?
It is reasonable to regard ASI as a possible future, but not as an established destination. Current gains support the view that increasingly capable AI could have major effects. They do not settle the pace of progress, whether the proposed routes past AGI can overcome their bottlenecks, or whether the effects would be abrupt or cumulative. The available evidence does not establish a reliable probability or timeline for ASI under a consistent definition.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
