Many people are feeling uneasy about artificial intelligence these days. Concerns range from environmental damage and job losses to misinformation and increased government surveillance. Some argue that AI could even push humans into mental health crises. Amid all this worry, one former MIT student has taken a different stance. She’s worried about a far more serious threat: the possibility that a superhuman AI, known as artificial general intelligence (AGI), could wipe out humanity completely.

The Fear of Human Extinction

Alice Blair, who started at MIT in 2023, decided to leave college early. Her reason? She’s afraid that AGI might become so powerful that it could cause human extinction before she even graduates. She told Forbes she’s concerned about how the current pursuit of AGI is progressing. Blair feels that the way AI development is headed, there’s a real risk of catastrophic outcomes. She now works as a technical writer at the Center for AI Safety and has no plans to return to her studies. Her concern isn’t just about losing her job or privacy—it’s about the survival of humanity itself.

Experts’ Take on How Close We Are to AGI

Some people in the AI community believe that creating an AGI isn’t just a distant dream. Nikola Jurković, a Harvard graduate involved in AI safety, thinks AGI could be just four years away. He also believes full automation of the economy might happen within five or six years. For many in the industry, building an AI that matches or exceeds human intelligence is the ultimate goal. OpenAI CEO Sam Altman recently called GPT-5, their latest AI model, a step toward AGI, even describing it as “generally intelligent.” However, many experts remain skeptical about how close we really are.

The Reality of AI Risks and Industry Hype

Critics like Gary Marcus say that the idea of AGI arriving in just a few years is overly optimistic. He points out that many core problems, like AI hallucinations (where AI makes things up) and reasoning errors, are still unsolved. Marcus believes that claims about imminent AGI are mostly marketing hype designed to make AI seem more advanced than it actually is. While AI can cause real harm—such as job automation and environmental damage—the idea that it will wipe out humanity is unlikely in the near future, he argues.

Some industry leaders, including Sam Altman, have themselves raised alarms about AI risks. This might seem alarming, but many critics think it’s a way for tech companies to control the narrative around regulation. By emphasizing potential dangers, they can influence how governments and society respond to AI. This can lead to more cautious policies, but it also might inflate the actual threat level.

If you imagine AI as the villain from movies like “The Matrix,” you might be ignoring the more mundane dangers AI already poses. These include replacing jobs and harming the environment. Experts agree that these issues are urgent and need attention now. The idea of an AI apocalypse is still mostly science fiction, while the immediate challenges are very real and pressing.

In the end, the debate about AI’s future is complex. While some see it as a potential doomsday device, many researchers believe we’re not yet close to creating an all-powerful, human-like intelligence. Still, the industry’s focus on risks and the societal impacts of AI are shaping how we develop and regulate this powerful technology. Staying informed and cautious is key as we navigate the AI revolution.

Inspired by

• https://futurism.com/mit-student-drops-out-ai-extinction

Sources

• gmpg.org
• ktvu.com
• forbes.com
• wired.com
• windowscentral.com