There are concerns about the viability and commercialization of generative AI, with delays in model releases and slow emergence of commercial applications. However, dismissing it as a bust overlooks the technology's real-world uses and potential. While commercialization takes time, it doesn't mean it won't happen eventually. There is a possibility that ultra-expensive models may not solve complex problems, leading to slower progress and fewer innovations. Nevertheless, this slower pace would still impact AI safety discussions, even though it shouldn't solely rely on recent AI hype. The fundamental case for AI safety rests on the potential dangers of developing powerful systems without proper oversight. Although the next generation of large language models may not pose immediate risks, overlooking safety concerns would be unwise. It is crucial to understand the intertwined nature of AI safety and AI hype. Even if the next-gen models aren't significantly better, AI will still transform society but at a slower pace.

Some skeptics believe superintelligence is possible but may require a different approach to address limitations. AI safety requires time to understand and solve its challenges. Public perception often links AI safety to the imminent arrival of superintelligent systems, but if they don't materialize soon, AI safety shouldn't be disregarded. It's important to consider the bigger picture and separate short-term investments from long-term societal implications. Regardless of specific AI advancements, safety will always be a crucial concern. We should approach powerful AI systems with caution and ensure their development is done safely. An unsuccessful attempt to create a dangerous system should prompt further policy considerations rather than dismissing the need for safety measures. It is essential to remain aware of the potential of AI while avoiding being swayed by hype or dismissing it altogether.