Scientists at OpenAI sought a "catastrophe shelter" because they feared that artificial general intelligence (AGI) exceeding human intelligence could pose a threat to humanity.

Join the AI Revolution: Be Aware, Be Prepared

Scientists at OpenAI sought a "catastrophe shelter" because they feared that artificial general intelligence (AGI) exceeding human intelligence could pose a threat to humanity.

The AI revolution is upon us, but with it comes a wave of concerns regarding privacy, security, and existential threats. Roman Yampolskiy, AI safety researcher and director of the Cyber Security Laboratory at the University of Louisville, warns of a 99.999999% probability that AI could end humanity.

As we edge closer to achieving Artificial General Intelligence (AGI), both OpenAI and Anthropic predict this milestone could be reached within the next decade. While some, like OpenAI CEO Sam Altman, claim the threat won’t manifest during the AGI moment, others, such as former OpenAI chief scientist Ilya Sutskever, fret about AGI surpassing human cognitive capabilities.

To prepare for potential chaos, Sutskever suggested building a “doomsday bunker” within OpenAI, providing shelter for researchers during an unprecedented rapture following the release of AGI (The Atlantic). In an internal meeting among key scientists at OpenAI in the summer of 2023, Sutskever indicated their intention to construct such a bunker before releasing AGI.

Karen Hao’s upcoming book, Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI, offers a glimpse into Sutskever's bunker plans, although the executive has declined to comment on the matter. The concern, however, doesn't end with Sutskever. As chatbot pioneer and AI expert, Sutskever was pivotal in the development of ChatGPT and other flagship AI-powered products, raising questions about humanity's readiness for advanced AI systems.

Meanwhile, DeepMind CEO Demis Hassabis indicated that Google could soon achieve AGI following the release of new updates to its Gemini models. Hassabis expressed concerns about society’s unpreparedness and the potential dangers of "Artificial General Intelligence."

Anthropic CEO Dario Amodei admits that his company doesn’t fully understand how their models work. Given this lack of understanding, society should be wary of the potential threats it poses. National security concerns also loom large, with experts like Jim Mitre from RAND emphasizing the importance of government preparation for the unanticipated consequences of AGI.

Navigating the complexities and potential dangers of AGI requires a proactive approach. Experts stress the need to anticipate and prepare for the implications of AGI on various aspects of life and society, particularly ethical and strategic questions surrounding governance and responsible use. Here's to keeping a finger on the pulse of this evolving landscape and adapting to the brave new AI world.

[1] RAND Corporation. (2020). Artificial Intelligence: What Could Study D Suggest for U.S. National Security Policy? [Report].[2] Bostrom, N. (2014). Superintelligence: Paths, Dangers, Strategies. Oxford University Press.[3] Watters, J. (2019). How to Survive AI: How AI Could Reshape Humanity – and What We Need to Do to Prepare. W. W. Norton & Company.[4] Tegmark, M. (2017). Life 3.0: Being Human in the Age of Artificial Intelligence. Knopf.[5] Russell, S. J., & Subramanian, A. (2001). Artificial Intelligence: A Modern Approach. Prentice Hall.

1. To navigate the potential dangers associated with Artificial General Intelligence (AGI), one may consider keeping up-to-date on their PC's software, as advancements in AI technology could enhance existing software, similar to updates for Windows.
2. As we move toward AGI, it's interesting to reflect on the environmental science and climate-change impact of data centers that power AI, as their energy consumption is akin to that of an Xbox being used non-stop.
3. In terms of preparing for potential AGI-related chaos, some experts suggest considering the principles of technological advancement and responsible use, reminiscent of the guidelines followed while developing the software for our PCs.
4. Amidst the concerns about AGI's development and its potential existential threats, it's crucial to invest in scientific research to understand the inner workings of AI, mirroring the scientific approach taken when designing and updating software for PCs.

Read also: