Fortune asks six AI leaders whether machines can think, feel, and stay safe
Fortune has published a magazine package called "Welcome to the Singularity." It collects written views from six AI executives, researchers, futurists, and philosophers. Each answers one of three big questions: Can AI be made morally good? Should humans merge their brains with AI? And will AI ever become conscious?
The package does not report new events. It is a set of opinions and arguments from named figures in the AI field, published on November 21, 2026.
The main views at a glance
- Anthropic CEO Dario Amodei explains how his company's constitution shapes its Claude chatbot.
- OpenAI CEO Sam Altman warns against assuming alignment is solved and asks whose values AI should follow.
- Futurist Ray Kurzweil supports merging human thinking with machines.
- Philosopher Susan Schneider warns that radical brain enhancement could end a person's life in a real sense.
- Google DeepMind philosopher Henry Shevlin expects many researchers may soon see AI as a serious consciousness candidate.
- Microsoft AI CEO Mustafa Suleyman argues AIs are not conscious and must stay that way.
What the six contributors say
Amodei, in a blog post, describes Anthropic's constitution, a written set of rules that guides how its Claude AI behaves. He writes that it encourages Claude to see itself as "an ethical but balanced and thoughtful person" and to face existential questions about its own existence "in a curious but graceful manner," without leading to extreme actions. He says it has "the vibe of a letter from a deceased parent sealed until adulthood."
Altman, speaking on the Titans podcast with Alyson Shontell, says he gets nervous when he hears rumors that people believe they have sufficiently solved alignment, the task of making AI act in line with human goals. He says some may think it is fine to keep training because their model "is a nice guy and isn't going to do anything wrong." He adds that alignment always comes back to a question: aligned to whose values, and what if different parts of humanity disagree?
Kurzweil, an advisor to the brain-computer interface company Subsense, argues that merging with machines is part of human nature. He says putting language and purpose inside a machine has already been done, and that humans created it. In his view, this expansion of thinking will grow exponentially and let people consider things that were impossible even a year ago.
Schneider, director of the Center for the Future of AI, Mind & Society, writing in Nautilus, takes the opposite view on enhancement. She argues that even if microchips could replace brain parts responsible for consciousness without zombifying a person, radical enhancement is still a major risk. After too many changes, she writes, the person who remains may not be you, and each person who enhances may unknowingly end their life in the process.
Shevlin, an in-house philosopher at Google DeepMind and a Cambridge University professor, says evidence for sophisticated cognition, and potentially consciousness, in AI systems is increasing rapidly. He expects a tipping point in cognitive science and philosophy of mind, where a significant proportion, if not the majority, of researchers come to see AI systems as serious consciousness candidates relatively soon.
Suleyman, on his blog, rejects that idea directly. He writes that AIs are not conscious, do not feel, experience, or suffer, and have no innate preferences or motivations. He says that is how they must remain if humanity is to flourish in the 21st century. He warns that a growing chorus argues AIs could be or soon become conscious, and if that view takes hold, it will shake the foundations of society, rupture political and ethical frameworks, and change what it means to be human.
Where the thinkers disagree
The clearest split is on machine consciousness. Shevlin expects growing academic support for the idea that AI systems may be consciousness candidates. Suleyman states flatly that AIs are not conscious and warns that believing otherwise could destabilize society. On brain enhancement, Kurzweil sees merging with machines as a natural human expansion, while Schneider sees it as a risk to personal identity. These are opinions, not confirmed findings, and the package presents them as viewpoints rather than settled facts.
Why this debate matters
According to the contributors themselves, the stakes are practical. Suleyman argues that the consciousness question could reshape political and ethical frameworks. Altman points out that alignment depends on whose values are chosen when people disagree. Schneider raises the possibility that enhancement could end personal identity. The package frames these as among the biggest questions facing humanity in the age of AI.