We live in a time when technological change is happening faster than ever. This sense of acceleration isn’t just a subjective impression—it’s a measurable reality. As early as 1999, Vint Cerf, one of the fathers of the internet, observed that one year in the internet industry was like seven “dog years.” That comparison, once apt at capturing the pace of innovation, now seems insufficient in the context of artificial intelligence. The speed at which AI is reshaping our world is unprecedented—faster than in previous technology waves, including the internet era. The amount of data and analysis on the topic is overwhelming, and media narratives often swing between utopian excitement and dystopian fear. Yet beneath the headlines lie hard numbers that paint a far more nuanced and fascinating picture.

In this article, I present five of the most surprising and counterintuitive findings from the latest analyses. They help explain the true nature of the AI revolution—its unprecedented speed, paradoxical economics, geopolitical tensions, impact on the physical world, and the fundamental shift in the labor market. These are truths worth knowing to navigate the era ahead with intent.

1. “Dog years” is an understatement: The speed of AI adoption defies belief

The pace at which AI-based tools have been adopted by users worldwide is unprecedented. The scale becomes clear when we compare how long different platforms needed to reach 100 million monthly active users or subscribers:

• Netflix: about 10 years to 100M global subscribers (2017)
• Facebook: ~4.5 years to 100M users (2008)
• Instagram: ~28 months to 100M MAU
• TikTok: ~9 months to 100M users
• ChatGPT: ~2 months to 100M MAU

Unlike previous waves, generative AI became a global phenomenon from day one. Lightning-fast adoption shortens feedback loops between users and builders, further accelerating innovation and increasing pressure to monetize from the very first day.

2. AI’s economic paradox: The more powerful it becomes, the stranger its business gets

The AI business model combines very high upfront investment with rapidly falling unit prices for services:

• Training costs are very high: estimates in model cards and academic publications show that emissions associated with training the newest, largest models reach thousands of tons of CO₂e (e.g., Llama 3.1 405B ~8,930 tons CO₂e per its model card). For older models such as GPT-3, academic papers report on the order of several hundred tons of CO₂e.
• Inference costs are dropping precipitously: since late 2022, prices for executing queries of comparable quality have fallen by more than 280×, dramatically lowering barriers to entry for builders and end users.

The result? AI is becoming cheaper and more accessible for developers and users, yet monetization for model creators is hard—they incur multibillion-dollar investments in technologies whose unit usage costs quickly fall.

It’s different this time, we’ll make it up on volume, and we’ll figure out how to monetize our users in the future are typically three of the biggest danger statements in business.

This is an editorial comment on business-model risks, not a conclusion drawn from the reports.

Hardware and accelerator performance

Inference cost declines are driven by scale, rising accelerator performance, increasingly effective model optimizations, and energy efficiency improvements across entire compute stacks.

Energy, water, and power grids

The scale of AI investment is forcing infrastructure expansion. Forecasts indicate that global electricity demand from data centers could double around 2026, and many countries are seeing growing pressure on water resources for cooling and on distribution grid build-out.

3. A new space race with two leaders and an open flank: Geopolitics and AI philosophy

Andrew Bosworth, CTO of Meta, likened the current AI phase to a “space race.” Today, competition runs on two fronts in parallel:

• Geopolitical front (U.S. vs. China): in 2024, the U.S. accounted for more “breakthrough” models than China (roughly 40 vs. 15), but the performance gap among top systems is shrinking quickly. New players such as DeepSeek are growing rapidly in their domestic market and increasingly assertive globally.
• Philosophical front (closed vs. open-weight): the quality advantage of closed systems has visibly narrowed—on Chatbot Arena the difference between leaders of both approaches has shrunk to ~1.7%.

These two fronts are intertwined. The decision to release strong models under an open-weight regime (e.g., Llama 3.1) is also strategic—it strengthens the global developer community and makes it harder for single vendors to monopolize the ecosystem.

Regulation: The AI Act and a global patchwork of law

The EU has adopted a comprehensive legal framework (the AI Act), which entered into force on 1 August 2024. Application is phased: some prohibitions apply sooner, while the full risk requirements come into effect progressively through 2026. Requirements for general-purpose models (GPAI) also enter into force under transitional arrangements.

Responsible AI and incidents

In parallel, the number of documented AI-related incidents is rising (errors, reputational harm, privacy violations). Organizations are implementing Responsible AI policies and model governance processes, but there remains a gap between declarations and actual enforcement.

Technical security: prompt injection, jailbreaks, and cyber

LLM-based systems are vulnerable to prompt injection and jailbreaks, among other threats. However, emerging best-practice catalogs and standards (e.g., the OWASP Top 10 for LLMs, NIST guidelines) help harden them (filters, sandboxing, output validation, vulnerability testing, defense in depth).

How we measure AI quality: Evaluations 2.0

Alongside classic benchmarks, the importance of factuality, robustness, and complex planning/agent tests is growing. At the same time, awareness of test-set contamination is rising, creating a need for credible, dynamic evaluation methods.

4. AI leaves the cloud: How AI is reshaping our physical world

The AI revolution isn’t happening in the cloud alone. The deepest change is the transformation of real-world assets—cars, agricultural machinery, defense systems—into intelligent software endpoints.

• Autonomous vehicles: the scale of miles driven is reaching the billions. Tesla’s FSD (supervised) surpassed about 4.5 billion miles in mid-2025. In robotaxi, Waymo already has a clear share of rides in San Francisco (about 22% in 2024), and independent analyses rate its safety as comparable to or better than the average human driver.
• Defense: Anduril is rapidly scaling autonomous defense systems (drones, surveillance, C2). Market estimates and press reports indicate that in 2024 the company doubled revenue to around $1B, reflecting rising demand for intelligent technologies in the defense sector.
• Agriculture: Carbon Robotics uses AI for laser “weeding,” which in practice can significantly reduce herbicide use (case studies cite savings of around ~80–90% in weeding costs) and improve efficiency.

Safety of autonomy on public roads

Analyses from pilot markets suggest that “driverless” systems may have lower collision rates than the average driver. Independent reviews classify Waymo as one of the “safest drivers” on available data, though methodology and scope remain subject to debate.

Agentic AI systems (agents)

Combining LLMs with planning, tools, and memory enables multi-step tasks in both digital and physical worlds. This opens new business models (process automation, customer support, fleet management) while creating fresh challenges in control, safety, and accountability.

AI in scientific discovery

“AI labs” that combine hypothesis generation, experiment design, and data analysis are shortening the discovery cycle in the natural sciences—confirmed by a growing number of peer-reviewed publications and real-world deployments.

5. It’s not whether AI will take your job—it’s whether you’ll learn to work with it

Cognitive automation is changing the nature of work. Labor-market data show a rapid rise in demand for AI skills: in 2024, the share of U.S. job postings requiring AI skills reached about 1.7–1.8% of all listings, and mentions of generative AI skills grew by more than threefold year over year.

“You’re not going to lose your job to AI, you’re going to lose your job to somebody who uses AI.” — Jensen Huang

Companies are adapting their practices. At Shopify, “reflexive use of AI” has become a baseline expectation for employees (disclosed in a publicly available internal memo). It’s a signal that working with AI is no longer a plus—it’s a basic competency.

The AI revolution does not so much eliminate the need for work as shift the center of gravity toward skills that amplify human creativity and judgment with computational power. The future belongs to those who treat AI not as a threat, but as the most powerful tool in their arsenal.

Conclusion

The facts presented here portray a revolution that is faster, more complex, and broader in scope than any other technological shift in history. Unprecedented global adoption pressures a paradoxical economics—driven by high training costs and falling inference prices. This dynamic intersects with geopolitical and philosophical rivalry for dominance, growing impact on the physical world, and a fundamental transformation of the labor market—key, interlocking pieces of the puzzle.

Understanding these phenomena is essential to move beyond simplistic narratives. AI is neither utopia nor apocalypse—it’s a powerful force already redefining the rules of the game in business, society, and international affairs. The challenge is to understand its dynamics and shape the future intentionally.

If AI is revolutionizing technology, business, and work at such breakneck speed, which fundamental truth about our society will it challenge next?

References and further reading

• Artificial Intelligence Index Report 2025 — Stanford HAI:
  report page · full PDF
• State of AI Report 2025 — Air Street / State of AI:
  online version · overview / note
• Trends – Artificial Intelligence (2025) — Mary Meeker / BOND:
  report page / PDF · slides on SlideShare