Expert Suggests AI’s Existential Threats May Be Overblown

The fast-paced development and adoption of artificial intelligence have raised concerns about its existential threats. However, Gary Marcus, an expert in the field, suggests that the risks may be overblown. Marcus, a New York University emeritus professor, has been cautioning against the rapid development of AI for the past six months. In a recent interview, he stated that the scenarios for AI’s extinction risk are not concrete enough to be a significant concern.

Building AI Systems We Don’t Have Control Over

Marcus is concerned about the risks posed by AI systems that we don’t have control over. He believes that we are building AI systems that we don’t have very good control over, and this poses a lot of risks. However, he doesn’t believe that the risks are existential, at least for now. This is because the scenarios for AI’s extinction risk are not concrete enough.

The Need for Regulation and Governance

There is a need for serious regulation and governance to ensure that AI is developed safely. Marcus argues that global leaders should focus on reducing the risk of extinction from artificial intelligence technology by creating national or international agencies responsible for AI governance. However, he warns against leaving the power to corporations. The big companies calling the shots here are not necessarily interested in the rest of us, he warns.

The Short-Term Risks to Democracy

In the short term, Marcus is worried about the risks posed to democracy by AI. He is concerned that generative AI software produces increasingly convincing fake photographs and videos at little cost. This poses a risk to democracy because elections may be won by people who are better at spreading disinformation. Moreover, democracy depends on having reasonable information and making good decisions. If nobody knows what to believe, then how do you even proceed with democracy?

Although Marcus is concerned about the risks posed by AI, he still sees a lot of upside. He believes that AI not yet invented can help with science, medicine, and elder care. However, in the short term, he feels like we’re just not ready. There’s going to be some harm along the way, and we really need to up our game. We have to figure out serious regulation and governance to ensure that AI is developed safely.