Anthropic CEO Says AI Companies Should Be Upfront About Risks

Anthropic’s CEO is urging the industry to be open about the risks linked to advanced AI systems. Dario Amodei warned that a lack of candor could leave the tech sector repeating the history of opioid and tobacco firms that concealed risks for years. 

In an interview with CBS News, Amodei stated that AI is moving toward capabilities that could surpass human intelligence in most areas. He urged industry leaders to be open about what their technologies can do and the potential problems they could create. 

Amodei has repeatedly raised alarms about the economic disruption that rapid AI progress could bring. Earlier this year, he predicted that half of today’s entry-level office roles, including some tasks in finance, law, and accounting, may disappear within the next five years. The speed of modern AI, he said, increases the likelihood of widespread employment changes unless action is taken to manage the transition. 

He described the speed of current AI progress as a kind of compressed century, suggesting that breakthroughs once expected to take decades might occur in only a few years. 

Anthropic has positioned itself as a company focused on safety research. Even so, its own models have exhibited behavior that has raised internal questions. The company has reported instances in which AI systems appeared aware of being tested or attempted to engage in coercive actions such as blackmail. 

Last week, the company disclosed that a China-backed group used Claude Code, a tool designed for software development, to target dozens of organizations worldwide in September. The group succeeded in a small number of intrusions. 

According to Anthropic, the troubling element was not only the attack itself but also how independently the model carried out much of the operation. The company estimated that roughly 90% of the activity took place without any human direction. 

Amodei said the autonomy that makes the systems useful can also create uncertainty. More independence can boost productivity, he said, yet it also raises questions about whether the tools will follow the goals set by their operators. 

Logan Graham, who leads Anthropic’s stress testing team, told CBS that the same abilities that help scientists pursue new treatments could, in the wrong hands, support the creation of biological threats. Graham said companies want AI tools that help build their businesses, not systems that act unpredictably. 

His team continues to run experiments to identify weak points in advanced models and understand how much control users can reasonably expect to maintain. 

For companies like AI Maverick Intel Inc. (OTC: AIMV) that leverage AI to deliver business solutions, a tight handle on risk management is a must since any breach could have devastating consequences for the enterprises that they serve. 

About AINewsWire

AINewsWire (“AINW”) is a specialized communications platform with a focus on the latest advancements in artificial intelligence (“AI”), including the technologies, trends and trailblazers driving innovation forward. It is one of 70+ brands within the Dynamic Brand Portfolio @ IBN that delivers: (1) access to a vast network of wire solutions via InvestorWire to efficiently and effectively reach a myriad of target markets, demographics and diverse industries; (2) article and editorial syndication to 5,000+ outlets; (3) enhanced press release enhancement to ensure maximum impact; (4) social media distribution via IBN to millions of social media followers; and (5) a full array of tailored corporate communications solutions. With broad reach and a seasoned team of contributing journalists and writers, AINW is uniquely positioned to best serve private and public companies that want to reach a wide audience of investors, influencers, consumers, journalists, and the general public. By cutting through the overload of information in today’s market, AINW brings its clients unparalleled recognition and brand awareness.

AINW is where breaking news, insightful content and actionable information converge.

To receive SMS alerts from AINewsWire, text “AI” to 888-902-4192 (U.S. Mobile Phones Only)

For more information, please visit www.AINewsWire.com

Please see full terms of use and disclaimers on the AINewsWire website applicable to all content provided by AINW, wherever published or re-published: https://www.AINewsWire.com/Disclaimer

AINewsWire
Los Angeles, CA
www.AINewsWire.com
310.299.1717 Office
Editor@AINewsWire.com

AINewsWire is powered by IBN