In 2026, Yuval Noah Harari warns that artificial intelligence is not just supporting human decisions, but is beginning to steer key social processes—from elections to wars. Are his theses about "hacking the code of civilization" justified, or is this alarmism? We examine the evidence, expert reactions, and real-world cases.
In May 2026, Yuval Noah Harari, author of bestsellers such as Homo Deus and 21 Lessons for the 21st Century, gave an interview to The Economist in which he compared artificial intelligence to a computer virus. In his view, AI is not so much assisting humanity as it is gradually taking control of the "operating system of civilization"—manipulating information, automating decisions, and altering the structure of power. Is this just a metaphor, or a real threat?
AI as a new "actor" in politics and media
Harari is not alone in his concerns. In March 2026, during the TED2026 conference, he presented specific examples where AI algorithms are already influencing key aspects of social life:
- Election manipulation: In January 2026, a deepfake campaign was detected in India, where AI generated false statements by politicians. Although the incident was quickly publicized, experts warn that similar attacks could become the norm in the era of digital disinformation (source: Reuters).
- Social media: TikTok and Instagram algorithms not only personalize content but actively create new political narratives. A 2026 MIT study found that these systems reinforce polarization by suggesting increasingly extremist content to users (source: MIT Sloan).
- Autonomous decisions: In October 2025, the New York Stock Exchange experienced a "flash crash" caused by AI algorithms that misinterpreted macroeconomic data. A Bank of England report from March 2026 warns that 70% of stock market transactions are initiated by AI, which increases the risk of instability (source: Bank of England).
Harari emphasizes that the problem is not just that AI makes decisions for us, but that we do so in a way we do not understand. In an interview with The Guardian, he stated: "It’s not that AI is smarter than us. It’s that it is faster, more unpredictable, and operates on a scale we cannot control."
Military, finance, medicine: Where is AI already making key decisions?
Harari's thesis about "hacking civilization" is taking real shape in several key areas:
1. Autonomous military systems
In 2025, the UN published a report on the use of autonomous drones in the conflict in Sudan. These systems made decisions about attacks without human intervention, which Harari called "the first case of AI as an independent war actor" (source: UNODA). Although this technology may reduce casualties among soldiers, it raises serious ethical questions: who is responsible for decisions made by machines?
2. Financial algorithms
In 2026, AI algorithms are responsible for the majority of transactions on global stock exchanges. As the 2025 "flash crash" showed, misinterpretations of data can lead to sharp drops in stock prices. Harari warns that the financial market has become too dependent on systems whose actions are not fully understood even by their creators.
3. Medical diagnostics
On the other hand, AI also brings benefits. In 2026, machine learning-based systems are helping in the fight against pandemics by predicting virus mutations (source: Nature). Harari does not deny these achievements, but stresses that it is crucial to maintain a balance between innovation and control.
Is Harari right? Expert reactions
Harari's theses have sparked a lively discussion among experts from various fields:
- Support:
- Geoffrey Hinton, Turing Award winner, stated in April 2026 that "Harari is right – AI is already making decisions we don't understand" (source: MIT Technology Review).
- Stuart Russell, professor of AI at UC Berkeley, supported Harari's call for global regulations, arguing that "without an international legal framework, we risk losing control over our own systems" (source: Wired).
- Criticism:
- Yann LeCun, Chief AI Scientist at Meta, accused Harari of "alarmism," claiming that AI is still a tool, not an independent actor (source: Twitter/X).
- Gary Marcus, cognitive scientist, believes Harari overestimates AI's capacity for manipulation, ignoring its limitations. In an article for Scientific American, he wrote: "AI is powerful, but it is not omnipotent. It still lacks true intelligence and an understanding of context" (source: Scientific American).
This discussion shows that the topic of AI evokes extreme emotions. Some see it as salvation, others as a threat to humanity. Harari positions himself somewhere in the middle – he does not deny the potential of AI, but warns against its uncontrolled development.
What solutions does Harari propose?
In the face of the growing role of AI, Harari calls for concrete actions:
- Global regulations: In June 2026, he participated in a UN panel on AI ethics, where he called for an international treaty regulating the development and application of artificial intelligence (source: UN News). In his view, similar to nuclear weapons, control mechanisms are needed to prevent abuse.
- Education: Harari emphasizes that it is crucial to prepare societies for life in the AI era. He proposes introducing mandatory classes on critical thinking and artificial intelligence in schools (source: BBC Future).
- Technological limitations: He suggests that certain areas, such as autonomous weapons or election algorithms, should be completely banned. He spoke about this during the Web Summit 2026, arguing that "some technologies are too dangerous to be left without control."
Harari is not optimistic, however, about the pace of these changes. In an interview with Politico Europe, he stated: "The problem is that technology is developing faster than our ability to regulate it. By the time we introduce appropriate regulations, it may already be too late."
AI in culture and media: How is the topic presented in 2026?
Harari's concerns are reflected in culture and media. In 2026, the topic of "AI hacking civilization" appears in many productions:
- Films and series:
- The Netflix documentary Hacked: The AI Takeover (premiered February 2026) analyzes cases of AI manipulation, featuring statements from Harari.
- The HBO series The Algorithm (2026) presents a fictional story about AI taking control of government systems.
- Books:
- The Code Breakers (2026) by Max Fisher, a journalist for The New York Times, describes the impact of AI on geopolitics.
- Homo Artificialis (2026) – Harari's new book, in which he expands on his thesis about "hacking civilization" (source: Penguin).
- Reports and articles:
- AI and The Future of Democracy – a report by the Oxford Internet Institute (April 2026).
- A special issue of The Atlantic, Can We Trust AI? (May 2026), featuring an essay by Harari.
These productions show that the topic of AI is no longer the domain of science fiction, but a real challenge we must face.
Summary: Is Harari right?
Yuval Noah Harari's theses about AI "hacking the code of civilization" are provocative, but they are based on real examples. In 2026, artificial intelligence is already:
- Manipulating information and influencing election results.
- Making key decisions in finance, medicine, and the military.
- Changing the structure of power, creating new narratives and alternative versions of reality.
Does this mean that AI is actually "hacking" civilization? The answer is not straightforward. On one hand, there is evidence that algorithms can manipulate human decisions. On the other, AI remains a tool whose actions depend on humans.
Harari does not deny the potential of AI, but warns against its uncontrolled development. His call for global regulations, education, and technological limitations seems reasonable, though difficult to implement. In the face of the growing role of artificial intelligence, it is crucial that we as a society learn to coexist with it – without allowing it to take control.
As Harari noted in one of his interviews: "It’s not about whether AI will destroy us. It’s about whether we will allow it to change us in ways we don't want."
Are we ready for this change? That question remains open.
If you are interested in the topic of AI and its impact on society, check out our other posts:
- Are algorithms ruling our elections? Mechanisms, companies, and the boundaries of digital power
- AI agents at work: How will they change the daily tasks of programmers and office workers in 2024–2026?
- Artificial Intelligence and the Labor Market: Revolution, Threats, and New Opportunities
Sources
- https://youtu.be/hBtVGwuJzpk?si=swa57TAuqJKRiDig
- https://www.economist.com
- https://www.ted.com
- https://www.theguardian.com
- https://www.reuters.com
- https://unoda-web.s3-accelerate.amazonaws.com
- https://www.sec.gov
- https://www.nature.com
- https://news.un.org
- https://www.politico.eu
- https://www.technologyreview.com
- https://twitter.com/ylecun
Comments