“AI is not just another tool; it is an agent. It can learn and change by itself and make decisions…” These were profound yet frightening words by Yuval Noah Harari when he spoke at Davos 2026 earlier this year in January.
Harari opened his speech with a sentence that shook the audience. His example of referring to an AI agent as a knife caught the audience’s attention. He said, “A knife is a tool. You can use a knife to cut salad or to murder someone, but it is your decision what to do with the knife. AI is a knife that can decide by itself whether to cut salad or to commit murder”.
The phrase ‘AI is a knife that can decide by itself’ must be contemplated by the gatekeepers and decision-makers governing the landscape of AI. We were taught, or have been living under the assumption, that humans decide while AI, software, or tools execute commands.
Harari’s words suggest otherwise: that tools or agents can decide for themselves. AI may not follow the rules humans make, but can create rules for itself. This heralds the rise of a dystopian era in which accountability and regulation may fade if AI takes over.
Harari then turned to a more philosophical question: what does it mean to be human in the age of AI? He argued that humans have long defined themselves by their ability to think, reason, and use language to create meaning. Yet AI is demonstrating capabilities that compete with humans in processing information and generating language. Harari issued a stark warning: if AI becomes better than humans at working with words, then any system built on language could be transformed by it.
‘Words’ were the key to Harari’s speech. AI’s ability to evaluate, examine, analyse, compile, argue, opine and present information, facts,and replies better than humans will create a crisis. It will be AI systems, tools, and agents that invade society by performing better than humans.
Harari asked, “Will your country recognise AI immigrants as legal persons?” – a question that stirred the audience’s thoughts. New Zealand granted legal rights, duties and liabilities to the Whanganui River in a historic ruling in 2017. However, unlike rivers or corporations, which are termed legal persons in many countries, AI can make decisions on its own. It can evolve and evaluate. AI bots have been active for years, conversing with humans to guide them.
Harari’s insights presented a key understanding of AI. If it becomes better than humans at generating and refining words and language, then humanity’s greatest asset may no longer belong exclusively to humans. For countries like Pakistan, where public opinion is increasingly influenced by digital platforms and social media, this question remains pertinent. The concern is not merely that machines may become intelligent, but that systems may become capable of producing persuasive content at an unprecedented scale to influence decisions.
What makes Harari’s AI immigration analogy particularly powerful is that it frames AI as something entering our social and economic systems whether we are prepared or not. Just as nations establish rules for refugees, societies must determine the conditions under which AI operates.
AI is not waiting for governments, businesses, or educational institutions to catch up. It is already transforming how information is created, shared and consumed. The real question is whether policymakers, regulators and citizens can develop effective safeguards before technological change overtakes their ability to respond.
The writer is a freelance contributor.
