The past three and a half years in AI have been intense, marked by hype, sector-specific disruptions, and deep transformations in how hundreds of millions of people work, write, create, and even think.
In a more existential, almost visceral sense, the past three and a half years have also been highly polarized.
Two almost opposite visions of the future have taken center stage: I will call them singularity-driven techno-optimism and humanism-driven techno-skepticism.
The AI industry promotes a view of an AI-enabled singularity that will create an intangible, almost indescribable abundance for everyone. There will be no diseases, no inconvenience, and infinite intelligence to solve all humanity's problems.
In fact, yesterday, Stripe's CEO announced in a letter to investors that the ‘singularity’ had arrived on January 1, 2026, and that Stripe was operating under this premise.
On the other side of this debate, holding a very different view of the future, there are people from various professional fields who are seeing the multi-layered disruptions caused by widespread AI deployment and are not so sure that this techno-optimistic dream will ever come true or yield truly beneficial outcomes.
They are also worried about systemic and existential risks, concentration of power, emerging harms, and the degradation of institutions that hold the social fabric together.
These two views of the future keep clashing (they often call each other ‘doomers’ and ‘accelerationists’), and these clashes make AI policy and regulation more complex and unproductive, as there is not enough common ground to foster initiatives that would truly benefit society.
I do not think these clashes are going to take us anywhere.
On the contrary, they will exacerbate hatred and division already rampant in many parts of the world for various reasons.
As a human collective, however, we can advance further if we seek common ground in our beliefs, values, and visions for the future.
AI is not going anywhere. Regardless of people's personal opinions, it is here to stay and will be part of our individual and collective identities for the foreseeable future.
We would be more productive, individually and collectively, if we focused on new post-AI utopias that would make anyone look forward to the future and feel like making plans and helping build them, regardless of their level of optimism or skepticism about AI or accelerated technological development in general.
It looks like, in AI, a vision of the future that excites different groups of people is one that focuses on specialized AI systems that directly benefit people and communities by solving specific problems.
For example, AI systems that look for new ways to diagnose or treat diseases, or that can help people live more dignified and happier lives.
Recently, Dario Amodei wrote:
“saying that AI will cure cancer is more a cliche than it is inspiring, and most people think it is deceptive. The thing that will work is actually curing cancer.”
I agree with that.
After learning more about many such initiatives and realizing that thousands of people are already working on these projects and substantial investments are being made in them, I have become much more optimistic about AI and how it could help people and society.
I have also become motivated to help create a future in which these initiatives are incentivized and AI governance and policy efforts, both directly and indirectly, support them.
I have been thinking a lot about what a post-AI future could look like and how we can bring people together, regardless of their views about AI or technological acceleration.
With that in mind, I would like to invite you to share more about your view of a post-AI utopia:
What makes you excited, optimistic, and looking forward to the future?
Are you involved in a specific ‘AI for good’ project? How can other people get involved?
What, for you, is a post-AI future that is worth fighting for?
I would love to hear your thoughts in the comments section below.
AI is a technology whose multitudinous applications across humanity cannot be categorised as good/optimistic or bad/pessimistic. It's like trying to determine whether electricity has been responsible for more good things than bad, it can clearly be used for either.
No. I reject AI.