Obama on agentic AI
Obama recently said: “If we are thinking about AI just in terms of how do we cure cancer or get better energy, you can do that without having agentic AI and having it just roaming free in the internet.” As Roon noted: “this sounds completely incoherent to me” From Rob Saker: An agent is a […] The post Obama on agentic AI appeared first on Marginal REVOLUTION .
Recently, former President Obama expressed his thoughts on artificial intelligence (AI), stating that focusing solely on curing ailments or improving energy without incorporating "agentic" AI, a system capable of planning, utilizing tools, writing code, querying data, conducting experiments, verifying its results, and repeating this process rapidly, is incomplete.
Roon, an unnamed individual, found Obama's statement "incoherent," while Rob Saker, another source, defined an "agent" as a system with the capability to perform various tasks autonomously.
Saker emphasized that merely treating "agentic" AI as an optional addition to a system, like a polite chatbot confined within a box, is a misguided perspective. He argued that curing diseases such as cancer necessitates systems that can comprehend existing literature, propose hypotheses, design experiments, analyze results, and iterate this process significantly faster than human researchers. Without agentic AI, one would merely have an "expensive autocomplete" without any practical value.
Obama's primary concern, according to Saker, is how to prevent commercial AI agents from operating freely. Possible solutions include implementing a Chinese-like firewall to block access to websites hosting foreign AI agents, closely monitoring these sites for potential threats, banning VPNs, granting government agencies the power to inspect computer hard drives, or introducing stringent CAPTCHA requirements on commercial websites.
However, Saker questioned how exactly to define an AI agent, noting that even the O3 system exhibited some agentic abilities, such as exploring museum websites.
Saker also pointed out the need for significant changes in cyber laws to accommodate the advent of agentic AI. He suggested that it would be an opportune moment to commission and cite peer-reviewed academic literature on the topic, as public discourse seems hesitant to address the freedom concerns, practicality, and civil liberties implications.
Lastly, Saker noted that Obama's remarks might have led to the bankruptcy of AI companies like Anthropic and OpenAI, but he did not provide further details on the reasons behind this speculation.
Written by urgent.news from Marginal Revolution's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
