On June 10, 2025, Sam Altman announced that humanity had crossed an invisible threshold. “We are past the event horizon; the takeoff has started,” the OpenAI chief executive wrote in an essay called The Gentle Singularity.

Altman did not claim that robots had taken over the streets or that a machine had suddenly become an all-powerful superintelligence. His argument was quieter and, in some ways, more consequential: the decisive transition had already begun even though ordinary life had not yet caught up with it.

AI agents, he predicted, would perform real cognitive work. Software development would never be the same. Scientific discovery would accelerate. AI would help researchers build better AI. Intelligence would become inexpensive, personalized and broadly available. Society would increasingly be limited by the quality of its ideas rather than by the difficulty of executing them.

A year later, Altman’s declaration deserves more than a benchmark scorecard.

Artificial intelligence has clearly improved. Models are more capable at coding, research assistance, document analysis, media generation and structured multi-step work. Used by a capable person, they can reduce the time required for many tasks and make some forms of expertise more accessible.

But the larger short-term revolution sold to companies, workers, investors and governments is much harder to find.

The broad productivity surge remains uneven. Sustainable enterprise returns remain difficult to establish. Organizations have removed employees in the name of AI while adding model costs, consultants, infrastructure, governance, human review and security exposure. Some are hiring in new areas or restoring capabilities they cut too early. Even where individual workers save time, companies have not consistently proven that those savings become greater output, stronger profits or better service after every new cost is counted.

Meanwhile, something more durable has happened.

The companies building AI have gained extraordinary access to capital and influence over public policy. Their expansion increasingly shapes decisions about electricity generation, transmission capacity, water, land, chips, cloud infrastructure, tax incentives and national security.

What Altman actually declared

Altman’s original essay blended observations, predictions and political argument. Some elements were measurable. Others were deliberately difficult to disprove.

He wrote that the “takeoff” had started and that humanity was close to building digital superintelligence. He expected 2025 to bring agents capable of real cognitive work and 2026 to bring systems capable of producing novel insights. By 2027, he expected robots to perform useful work in the physical world.

He also described the industry as building “a brain for the world”—a system that would be easy for everyone to use and would make intelligence abundant.

The phrase “past the event horizon” carried enormous rhetorical force because it framed AI development as irreversible. If the takeoff had already begun, delay became dangerous. Caution could be portrayed as surrender. Governments and businesses were no longer choosing whether to reorganize around AI; they were choosing whether to keep up.

That framing matters because it helped turn uncertain forecasts into urgent investment decisions.

What clearly improved

A credible examination must begin by acknowledging what changed.

AI-assisted coding became more useful. Models became better at navigating files, editing code, calling tools and completing structured workflows. Research systems became more capable at searching large bodies of information, proposing lines of inquiry and assisting specialists. Creative tools improved in music, voice, image and video generation. Translation, accessibility and document processing became faster and less expensive.

For an independent creator, a small company or a researcher without institutional resources, these gains can be extraordinary. Tasks that once required a team can sometimes be prototyped by one person.

But a capability improvement is not the same thing as an economic transformation.

AI has repeatedly demonstrated the first category. Evidence becomes more complicated as we move toward the third.

A faster draft may still require extensive verification. More generated code may create a larger review burden. Automated customer responses may reduce first-line handling while increasing difficult escalations. A worker may finish a task earlier without the company converting the saved time into additional production.

This is not failure. It is a warning against treating a local efficiency gain as proof of a company-wide productivity revolution.

The promised productivity revolution remains difficult to prove

Worldwide AI spending is accelerating much faster than clear evidence of enterprise return.

Gartner forecast in May 2026 that worldwide AI spending would reach approximately US$2.59 trillion during the year, a 47% increase over 2025. The firm said the market was being dominated by vendors and hyperscalers while enterprises had not yet fully translated AI demand into their own spending and value creation.

Spending at that scale proves that companies believe AI is strategically important. It proves competitive fear. It proves that technology providers are building capacity.

It does not prove that customers are receiving proportional economic value.

The central accounting problem is that organizations often measure the visible benefit while distributing the costs across several budgets. A department reports time saved. Information technology pays for integration. Legal reviews the contracts. Security monitors data exposure. Managers supervise adoption. Senior staff verify output. Finance struggles to connect the entire system to revenue or profit.

The monthly model subscription is therefore not the cost of enterprise AI. It is the admission price.

The real AI budget can include:

• cloud computing and model usage;
• data preparation, retrieval and storage;
• application development and integration;
• consultants and implementation partners;
• identity, permission and access controls;
• evaluation, monitoring and quality assurance;
• employee training and workflow redesign;
• legal, privacy and regulatory review;
• cybersecurity controls and incident response;
• human verification, correction and escalation.

These investments may ultimately be worthwhile. The point is that they must be counted before an organization declares victory.

Layoffs were treated as proof before the proof existed

The most aggressive short-term AI promise was not that workers would receive better tools. It was that companies would need substantially fewer workers.

Executives described flatter organizations, automated administration and permanently smaller teams. Investors often rewarded restructuring announcements as evidence that a company had become “AI-first.”

But downsizing is not the same as productivity.

A company can eliminate thousands of jobs and lower payroll without demonstrating that AI performed the missing work. It may have cancelled projects, accepted slower service, transferred work to customers, reduced quality, increased pressure on remaining staff or shifted responsibilities to contractors.

Adecco Group chief executive Denis Machuel said in July 2026 that companies were sometimes blaming AI for job reductions that were actually driven by restructuring or underperformance. His warning was especially important for entry-level work: eliminating too many junior roles may damage the pipeline that produces future experienced workers.

At the same time, Reuters has documented companies explicitly linking cuts and resource shifts to AI. The layoffs are not imaginary. The unsupported leap is assuming that every removed employee represents a successfully automated function.

Gartner’s more revealing finding was that workforce reductions may create room in a budget without creating a return. Reporting on its research indicated that organizations cutting jobs alongside autonomous AI deployments did not show a significant relationship between those layoffs and stronger AI ROI. The organizations seeing better returns were more likely to invest in human skills, governance and redesigned roles.

That produces a very different interpretation of the moment:

Before accepting a layoff as evidence of AI success, investors, journalists and workers should ask:

• What measurable output increased after the reduction?
• What happened to service quality, errors and customer satisfaction?
• Were consultants, contractors or specialist hires added elsewhere?
• How much did model access, infrastructure and security cost?
• How much human review remained necessary?
• Did the company later restore any lost capability?

Without those answers, a layoff proves that workers were removed. It does not prove that AI replaced them successfully.

The work did not disappear. It moved.

Generative AI arrived through an interface that concealed complexity.

Type an instruction. Receive an answer.

That experience made automation appear cleaner than it is. A dependable organizational system requires accurate data, controlled access, reliable integrations, monitoring, testing, escalation and accountability when the system fails.

It also needs people who understand when the answer is incomplete, misleading or dangerous.

AI can therefore relocate work rather than eliminate it:

The visible production task becomes cheaper. The surrounding system becomes more demanding.

This is one reason some organizations cut people and later hire in adjacent roles. The original task changed, but the need for judgment, exception handling and accountability remained.

Security risk belongs inside the ROI calculation

Security was too often treated as a later implementation issue rather than part of the original business case.

AI systems increase the number of ways sensitive information can move through an organization. Employees may paste confidential material into unapproved services. Generated code may contain vulnerabilities. Connected agents may receive excessive permissions. Prompt-injection attacks can manipulate systems into revealing information or taking unintended actions. Retrieval tools can expose the wrong data to the wrong person.

The World Economic Forum’s Global Cybersecurity Outlook 2026 found that 87% of respondents identified AI-related vulnerabilities as the fastest-growing cyber risk during 2025. Data leakage connected to generative AI and increasingly capable adversaries were among the leading concerns for 2026.

The same technology can strengthen defence, automate detection and reduce incident costs when deployed responsibly. That does not erase the new exposure. It reinforces the need to count security architecture as part of the deployment.

One serious failure can wipe out years of modest efficiency gains.

The clearest AI achievement may be political

While businesses struggle to prove consistent returns, the AI industry has already accomplished something more durable.

It has convinced governments that advanced computing capacity is a strategic national requirement.

That claim is not entirely hype. Countries without dependable access to advanced computing could become dependent on foreign cloud platforms and foreign models. Domestic capacity may matter for scientific research, healthcare, defence, public administration, cybersecurity and control of sensitive data.

But once AI is framed as an unavoidable national competition, the companies building it gain leverage.

The argument becomes:

This is where Altman’s “event horizon” matters beyond philosophy. The language of inevitability can turn private forecasts into public urgency.

Governments may then make long-term resource decisions before the promised short-term economic benefits have been independently demonstrated.

The race to control AI begins with the data centre

Artificial intelligence does not exist in a weightless cloud.

It depends on buildings filled with processors, networking equipment, storage, cooling systems and backup power. Those facilities depend on electricity, water, fibre, land, financing, skilled labour and government approval.

That physical reality is the foundation of my series, Who Controls the Intelligence?

The series asks a question that the product conversation regularly avoids: when advanced intelligence becomes dependent on enormous physical systems, who controls the resources required to build and operate them—and who carries the risk?

Data centres are becoming requirements for the future the industry has promised. That increases the influence of the companies able to build them and the governments willing to support them.

A large operator does not simply buy electricity after opening a building. Its projected demand can influence decisions about new generation, substations, transmission lines, reserved grid capacity and who carries the cost if the project changes.

The same is true of water and land. Resource access becomes a negotiation over priority.

The central issue is not whether AI data centres should exist. They will be essential to many useful services and scientific advances.

The issue is whether private companies receive lasting control over strategic infrastructure while the public absorbs a disproportionate share of the cost, risk and resource pressure.

Public money can reduce private risk

Governments have legitimate reasons to invest in computing capacity. Public infrastructure can support universities, hospitals, smaller companies, researchers and government services. Domestic capacity can reduce dependence on foreign suppliers.

But public participation must produce a measurable and enforceable public return.

Support can take many forms:

• tax exemptions and abatements;
• discounted or specially negotiated electricity;
• new substations and transmission infrastructure;
• roads, water systems and prepared land;
• grants, loans and financing guarantees;
• accelerated approvals and reserved capacity;
• training programs and public procurement commitments.

Some of those investments can produce lasting value. A stronger grid, new fibre connection or public research allocation may benefit a community beyond one corporate tenant.

But the largest number in a project announcement usually describes what the company may invest. It does not automatically reveal what the public contributes, who retains the asset or what happens if demand falls short.

That question becomes more urgent if the near-term AI business case has been overstated.

If revenue and productivity gains arrive more slowly than predicted, taxpayers and ratepayers may remain committed to infrastructure built around forecasts they did not create. Private operators may still retain valuable land, power agreements, buildings and strategic position.

An AI bust could strengthen the largest companies

The phrase “AI bust” can create the wrong picture. It suggests that the technology disappears and the winners lose their power.

A correction could produce the opposite result.

Smaller AI developers face high model costs, cloud commitments, expensive talent and constant pressure to finance the next generation of systems. Many applications remain easy for larger platforms to copy or bundle. If investor patience declines, smaller firms may fail, sell themselves or become dependent on the cloud companies they were expected to challenge.

The largest technology companies can spread infrastructure spending across profitable businesses in cloud computing, advertising, commerce, consumer devices and enterprise software. They can purchase distressed assets, hire displaced researchers and negotiate power contracts unavailable to smaller competitors.

The AI boom could therefore disappoint economically while succeeding politically and structurally.

The industry’s promises may weaken while its control over the chokepoints becomes harder to escape.

Cheap access is not democratic control

Altman predicted that intelligence would become inexpensive and widely available. At the user interface, that prediction may be partly correct.

Millions of people can now use capabilities once reserved for large institutions. A creator can produce a song demo, visual concept, marketing draft, software prototype or research plan at remarkably low cost.

But access is not ownership.

The company providing the model can change the price, generation limits, permitted uses, commercial terms, content policy, export options or account status. It can discontinue the model. It can alter how customer work is stored or processed. It can remove a feature on which a business depends.

Intelligence can feel abundant while the infrastructure underneath remains concentrated.

This is not democratized control. It is mass access to privately governed capability.

What this means for independent creators

Creators should not reject AI because the corporate story was overstated.

The tools can be genuinely empowering. They can help independent people test ideas, overcome technical barriers and create work that would previously have required larger budgets or specialist teams.

But the first year after Altman’s declaration offers a clear warning: do not build a creative business on the assumption that access, price and platform policy are permanent.

The people most exposed to an AI correction may not be those who refused to use the technology.

They may be those who built everything on the belief that today’s platform, pricing and permissions would remain unchanged.

One year later: the verdict

Sam Altman may ultimately prove correct about the long arc of artificial intelligence.

AI could accelerate medicine, scientific research, accessibility, engineering, education and creative production. We may achieve extraordinary advances across many fields faster than at any previous point in human history.

That possibility should not be confused with proof that the immediate commercial story was accurate.

Future breakthroughs do not establish that companies were right to remove workers before dependable replacements existed. They do not prove that current AI spending will produce proportional returns. They do not erase integration and security costs. They do not prove that communities should accept open-ended resource demands. They do not justify giving private companies lasting influence before the public receives measurable value.

One year after Altman declared that the singularity had begun, the clearest transformation is not yet a world overflowing with broadly shared productivity and prosperity.

It is the construction of a new infrastructure order.

The companies at its centre are gaining influence over chips, cloud capacity, electricity, land, water and public policy—resources that will remain important whether the current generation of AI products succeeds, consolidates or disappoints.

Sources and verification notes

This feature distinguishes verified developments from forecasts, corporate claims and analysis. Figures should be reviewed again immediately before publication if the article is published after July 27, 2026.

1. Sam Altman, “The Gentle Singularity,” June 10, 2025: blog.samaltman.com/the-gentle-singularity
2. Reuters, “AI will not trigger employment collapse, staffing company Adecco Group says,” July 23, 2026: Reuters report
3. Reuters, “Companies cutting jobs as investments shift toward AI,” updated July 6, 2026: Reuters tracker
4. Gartner, worldwide AI spending forecast, May 19, 2026: Gartner forecast
5. World Economic Forum, Global Cybersecurity Outlook 2026: WEF report
6. Jack Righteous, “The Race to Control AI Begins With the Data Centre”: Part 1
7. Jack Righteous, “Who Pays to Power the AI Race?”: Part 2
8. Jack Righteous, “The Water Behind Artificial Intelligence”: Part 3
9. Jack Righteous, “Public Money, Private Computing Power”: Part 4