OpenAI President Greg Brockman told the WSJ Tech conference that death could become a choice and predicted meaningful progress in treating disease within the next 12–18 months

OpenAI President Greg Brockman told the WSJ Tech conference that death could become a choice and predicted meaningful progress in treating disease within the next 12–18 months

At The Wall Street Journal’s WSJ Tech conference on November 4, Brockman was asked how superintelligent AI would change life for other species. He replied that “we all suffer” and that many things were possible, including making death a choice.

Near the end of the conversation, Brockman was told that many AI safety researchers put the probability of AI killing everyone within ten years at more than ten percent. He replied that it was difficult to draw firm conclusions from such numbers. The next question concerned other species. Brockman replied:

“If you take a field like biology, it extends beyond humans. We all suffer, and even considering the possibility that death could become a choice… all of this is possible. Cures for diseases have become a familiar promise in our industry, but I think that in the next 12–18 months you will see real results, meaningful progress and an impact on biology, health and the treatment of disease. It will take time to see the full effect, given the nature of the field.”

He went on to say that AI models were solving previously unsolved problems and that “we need to apply this to biology and the life sciences.” Earlier in the conversation, he gave two examples. First, he said that “fleets of agents,” large groups of AI programs, had solved a problem involving the Navier-Stokes equations that had resisted some of humanity’s brightest minds. The problem asks whether the velocity of a smoothly flowing fluid can grow without bound in a finite time. On September 8, OpenAI announced that it can, if the fluid starts at rest and is driven by a suitably chosen external force. OpenAI published the second example on the day of his appearance: a report describing 948 vulnerabilities in the Linux operating system kernel, with working exploits that give an ordinary user administrator privileges. Brockman said that around twenty such vulnerabilities are typically found in the kernel each year.

Brockman discussed the uncertainties in biology on the October 15 episode of the No Priors podcast. He attributed AI’s success in mathematics primarily to the abundance of data. For biology, he said, “the question, which remains unresolved, is whether we have the data and whether we can obtain it.” The OpenAI Foundation, OpenAI’s parent organization, is funding such data: in September, it allocated more than 125 million dollars to open data for biology and medicine. He also identified regulatory paperwork as a bottleneck: preparing the justification for a clinical trial takes about a year, and AI can draft these documents much more readily.

TechSpot and Wccftech led their headlines with his remarks about death. TechSpot noted that promises of breakthroughs within the next 12–18 months have been made since ChatGPT launched.

Wccftech compared Brockman’s remarks with Elon Musk’s statement on the January 6 episode of the Moonshots podcast, where he called longevity, or semi-immortality, “an extremely solvable problem.” Musk’s reasoning was that the body ages in sync: “Nobody has an old left arm and a young right arm,” so the clocks governing aging should be readily identifiable. He concluded: “You are programmed to die, and if you change the program, you will live longer.”

Immunologist Derya Unutmaz responded to a video clip of Brockman’s answer by reiterating a timeline he said he had proposed more than twenty years ago: longevity escape velocity (the point at which medicine adds more than a year of life for every year lived) would be reached around 2035, after which aging would be reversed and “death would become a choice.” His conservative estimate was the 2040s.