AI is a rapidly developing technology that has the potential to revolutionize many aspects of our lives. What could potentially be an incredible advancement, however, could also carry high risks to the future of humanity. Represented in the story of Edward Teller’s research with nuclear weapons, the dangers of AI cannot be underestimated.
The comparison between AI and gain-of-function research is helpful in understanding the gravity of the situation, but does not capture the motivations behind this technological advancement, which are more in line with intentions behind social media companies. Whereas experiment virology is mostly carried out by university professors and has the goal of protecting people more than turning a profit, those behind AI seek to make money from the advancement of the technology, which has high barriers to entry and comes with many global geopolitical dynamics.
Although social media has had damaging effects, it does not present an existential threat like AI does. AI comparison is thus more accurate when compared to nuclear weapons in terms of the potential destruction that can be inflicted through it, and the tension between countries.
On the other hand, AI is also like electricity due to its widespread applicability and its lack of specificity. Electricity has not posed an imminent threat to humanity, however, while AI may carry the risk of humanity’s extinction through its superintelligence one day. To more accurately represent the nature of this threat, it is more accurate to make an analogy between AI and Edward Teller’s fear of atmospheric ignition with nuclear bombs – in other words, AI presents the possibility of a destructive force that is unleashed and cannot be reversed.
Anthropic Technologies is a company that has partnered with Google to develop artificial intelligence. It was founded in 2017, initially as a startup providing crowdsourced data solutions, and since 2018 it has focused solely on advancing AI technology. Anthropic engineers and data scientists work to create trust-worthy AI algorithms and models. Four of the top Anthropic AI researchers recently quit the company to start their own venture, which was valued at $100 million.
Edward Teller was an American nuclear physicist. He was heavily involved in the Manhattan Project and warned his colleagues that detonating the first nuclear bomb may lead to the end of the world through atmospheric ignition. He ran calculations to assess the risks and advocated for caution based on his findings, leading to his colleagues seriously considering his message. Ultimately, his concerns were proven unfounded and the project continued. His story of caution stands as an analogy for the moment of AI today where risk and promise walk hand in hand.
As a rapidly advancing technology, AI has the potential to be incredibly beneficial for humanity, but it could also have the power to completely upend our way of life. AI is like gain-of-function research, social media, nuclear weapons, electricity, and atmospheric ignition – all in one. As we assess and react to this new technology, we should pay heed to Teller’s story as a reminder of the gravity of this technological development and the threat it may pose.
