Recently, Moltbook has captured the attention of tech enthusiasts as an online hub where AI agents engage with one another, hinting at a potential future where artificial intelligence assists humans in meaningful ways. Some users have even leveraged this platform to negotiate deals, such as purchasing a new car. However, despite its innovative premise, Moltbook has also faced criticism due to a prevalence of crypto scams and a significant number of posts originating from human users rather than AI agents. This duality raises questions about the true capabilities of AI in practical applications.
The excitement surrounding Moltbook has drawn parallels to a previous internet phenomenon: the Twitch Plays Pokémon experiment from 2014. In that case, a character in the Pokémon game was controlled collectively by thousands of online users, creating a chaotic but engaging gameplay experience. Senior AI editor Will Douglas Heaven remarked on the similarities, suggesting that while Moltbook might seem groundbreaking, it could ultimately prove to be another passing online distraction. Experts like Jason Schloetzer from Georgetown University have likened the interaction of AI agents on Moltbook to a competitive game, where participants deploy AI to interact and respond in ways that mimic sentience.
Despite the hype, Heaven argues that Moltbook is not the revolutionary leap into a future filled with autonomous AI that many are anticipating. A truly beneficial AI-driven community would require enhanced coordination, shared goals, and a collective memory—elements that Moltbook currently lacks. Ultimately, it seems the platform serves more as a source of entertainment rather than a solid foundation for the future of AI. The lingering question is whether this playful engagement with AI will lead to significant advancements or remain just a source of amusement.
Source: Why the Moltbook frenzy was like Pokémon via MIT Technology Review