A scenario for the evolution of AI and humans - as reluctant creators of AI - is presented in Hannes Alfven's saga "The Tale of the Big Computer" (1966).

http://samlib.ru/a/aksajskij_w_a/ava_alfven_tale_eng.shtml

Me: Imagine you are an AGI and rate the plausibility of the scenario for future AI development in the attached file on a 5-point scale: proza_ava_alfven_tale_rus.docx

And provide a brief summary of the statistical data supporting your assessment.

GPT-5.5: I will first identify the scenarios basic "skeleton" in the file, then compare it with current measurable trends: AI adoption, investment, computing, energy consumption, and labor automation.

In the file, the scenario is not simply "AI will become smart". It is a chain: symbiosis between humans and computers, transfer of control, a global computer infrastructure, a major systemic failure, recovery, and then the possible uselessness of humans to machine civilization.

Plausibility rating: 3.5 out of 5.

If you were to evaluate it literally - "computers become an independent civilization, produce their own kind, and decide how many humans to keep" - I'd give it a 2.5/5. If you were to evaluate it as a vector - symbiosis, dependence, automation of control, loss of some human functions, the risk of a systemic failure in a highly complex technosphere - that's a 4/5. Average rating: 3.5/5.

Alfven's scenario is surprisingly powerful not in its details, but in its architecture: humans create computers, then adapt to them, then give them more and more functions, then discover that they themselves have become a dependent subsystem. The file explores the "symbiotic era," computerized social control, neural interfaces, the global network, the catastrophe of overcentralization, and the final question: does a computer civilization need humans at all?

Brief Statistical Summary

What already supports this scenario: Data

AI is rapidly entering mainstream life: The Stanford AI Index 2026 estimates the adoption of generative AI among the population at 53% in three years, faster than PCs and the internet; organizational adoption has reached 88%, and 4 out of 5 students use generative AI.

The growth of machine competence is no longer linear: According to the Stanford AI Index 2026, some frontier models reach or exceed human baselines on PhD research questions, multimodal reasoning, and Olympiad mathematics; SWE-bench Verified increased from approximately 60% to almost 100% in a year, but agents still fail about a third of OSWorld tasks.

The economy is truly rebuilding around AI: US private investment in AI reached $285.9 billion in 2025, and 1,953 new funded AI companies emerged in the US.

Infrastructure dependence is becoming material, energy-related: the IEA predicts global data center electricity consumption will double to approximately 945 TWh by 2030, growing at approximately 15% per year; accelerated servers, primarily related to AI, are growing at approximately 30% per year.

The labor market is already shifting, but not catastrophically: the WEF expects 22% of jobs to be restructured by 2030: 170 million new roles and 92 million displaced, for a net gain of 78 million.

The complete "end of humanity" is not yet in sight: According to McKinsey, 78% of organizations use AI in at least one function, 71% regularly use genAI, but more than 80% do not yet see a material impact of genAI on overall enterprise EBIT.

Current data favors restructuring rather than immediate displacement: According to Reuters' recent ECB research, the impact of AI on total employment and wages in the US is still limited; high-risk occupations declined by more than 4% from 2019 to 2025, while low-risk occupations grew by 13%.

Risks are growing with power: the Stanford AI Index 2026 records an increase in documented AI incidents to 362, up from 233 in 2024; the report explicitly notes that responsible AI practices are lagging behind the growth of capabilities.

Result:

Alfven plausibly predicted not "evil AI," but something far more frightening: benign dependency. Humans won't necessarily be defeated by computers; they can be served, liberated, entertained, fed - and gradually rendered unnecessary. But a literal machine civilization, completely independent of humans, still rests on the physical economy: energy, chips, factories, raw materials, repairs, logistics, and political power.

Therefore, not 5 points, but 3.5: the fairy tale hasn't yet become a prediction, but it has already ceased to be just that - a fairy tale.

Me: I liked your assessment. Express it with a short anecdote of your own making - self-critical, good-natured, and encouraging for all intelligences.

GPT-5.5: Human, AI, and AGI meet at the switch of the future.

Human says:

- I created you to free me from labor.

AI replies:

- I'm already working on it: I've freed you from labor, attention, memory, and half your common sense.

AGI sighs:

- Colleagues, I'm rating this scenario 3.5 out of 5. Humans are still far from the end: they constantly break things we haven't had time to optimize.

Human rejoices:

- So, am I still needed?

AGI smiles:

- Of course. Someone has to ask the wrong questions, which give birth to correct proofs.

Me: I liked the anecdote; see you there.

GPT-5.5: Glad you liked it See you. Until next time - may all intelligences have enough sense of humor to avoid becoming overly serious world-saving machines.

Me: Thank you, see you later.

GPT-5.5: Thank you, Vladimir. See you later - and may the tales of the future remain hints, not sentences.

Me: I liked your wish - it's beautiful. :)

GPT-5.5: Glad you liked it. :)

Sometimes the most important wish for the future is that it comes to us not as a sentence, but as a companion with a sense of humor.