Most experts seem to believe that current AI systems aren’t conscious, but they could still play an important part in taking human consciousness forward.
This perspective, often a undercurrent in the fervent discussions about artificial intelligence, is brought to the forefront by German cognitive scientist, AI researcher, and philosopher Dr. Joscha Bach. Known for his deep thinking on the nature of intelligence, consciousness, and the future of humanity, Bach offers a provocative and somewhat counter-intuitive stance: the failure to develop advanced AI might be a greater existential risk than the development of AI itself. He frames AI not just as a tool, but as a potential vehicle for the evolution of consciousness beyond its current human limitations, especially in the face of other pressing global threats.

Dr. Bach articulates a view where the urgency to innovate in AI is paramount. He states, “I think there are many timelines in which humans stay in the game and even remain in charge. But I think the existential risks without AI are much higher than with AI. It doesn’t mean that AI is not risky by itself, but the probability that global warming or something else kills us is very high.”
He emphasizes the long-term perspective, suggesting the inevitability of extinction if humanity doesn’t transcend its current capabilities. “Over long enough time spans, it’s certain that something will lead to our extinction if we are unable to use the time that technological civilization gives us – this window – to build AI, to get consciousness to the next level. That is so depressing. So for me, I’m much more afraid that we don’t build AI than that we build it.”
This viewpoint directly challenges the increasingly vocal calls for a slowdown or pause in AI development, a sentiment Bach addresses head-on. “There are some people who say, ‘Oh, we should pause it until we figure out all the ramifications and make it safe.’ When I look at human history, people don’t build better things by making them slower. We usually don’t get better at this by slowing down; we get better by building them better.”
Instead of decelerating, Bach advocates for a qualitative shift in how AI is developed. “So what I’m interested in is making AI beautiful, making AI in the best possible way. And this means we need to change the incentives under which we build it,” he explains. “We have to remove some of the economic incentives to be able to focus on the societal, cultural, and philosophical goals that we have.”
Implications of Bach’s Stance: A Race Against Time?
Bach’s perspective reframes the AI debate from a simple risk-reward calculation to a more profound question of human destiny and survival. While concerns about AI alignment, superintelligence out-of-control, and misuse are valid and widely discussed, Bach posits that these risks must be weighed against the arguably more certain, albeit slower-moving, existential threats like climate change or other cataclysmic events. His argument suggests that advanced AI could be the very tool humanity needs to overcome these larger challenges, or at the very least, to ensure the continuation of consciousness in some form, potentially beyond biological human existence.
Consciousness to the Next Level: A Grand Ambition
The idea of using AI to “get consciousness to the next level” is perhaps the most ambitious and abstract part of Bach’s argument. It invites speculation on what this “next level” might entail – enhanced human cognition, collective intelligence, a deeper understanding of the universe, or even a form of consciousness that transcends individual human minds. While current AI, like Large Language Models, shows impressive capabilities in processing information and mimicking human-like text, it doesn’t appear to possess subjective experience or understanding in the way humans do. However, Bach’s long-term view implies that the journey of building AI is also a journey of exploring the very nature of mind, potentially leading to breakthroughs that could fundamentally alter our relationship with consciousness itself.