403
Sorry!!
Error! We're sorry, but the page you were looking for doesn't exist.
Powerful AI Model Kept Restricted as Safety Concerns Mount
(MENAFN) A US-based artificial intelligence company has reportedly developed a highly advanced model that it considers too risky for public release, according to statements from its executives cited in reports.
The company confirmed that it is currently engaging with US authorities to evaluate the capabilities and implications of the system, known as Claude Mythos Preview.
This development follows recent tensions between the firm and the US government. According to reports, US President Donald Trump previously imposed a temporary ban on the use of the company’s AI tools across federal agencies, citing concerns about national security and alleged pressure on defense institutions. During the same period, the Department of Defense reportedly partnered with a competing AI firm, OpenAI, to deploy its technology in sensitive military applications.
The company has also faced internal data exposure incidents. Earlier in the year, confidential documents related to the unreleased model were unintentionally made accessible after being stored in a public data cache. In a separate case, more than 500,000 lines of proprietary code tied to its AI systems were accidentally published online. The firm described the incident as “human error, not a security breach.”
According to statements from a senior safety official cited in reports, the new model demonstrates an unprecedented level of independence and analytical capability. It was described as “extremely autonomous” and capable of functioning similarly to a highly skilled cybersecurity expert, including identifying vast numbers of system vulnerabilities and generating corresponding exploit methods—an ability not seen in earlier versions.
In additional remarks cited in reports, the same official emphasized the broader significance of the development, stating that it represents “the starting point for what we think will be an industry change point, or reckoning, with what needs to happen now.”
The company has indicated that access to the model will be tightly controlled, limiting its availability to a small group of technology and cybersecurity organizations. According to its official communications, a wider release will only be considered once sufficient safeguards are implemented to mitigate the risks associated with its advanced capabilities.
The company confirmed that it is currently engaging with US authorities to evaluate the capabilities and implications of the system, known as Claude Mythos Preview.
This development follows recent tensions between the firm and the US government. According to reports, US President Donald Trump previously imposed a temporary ban on the use of the company’s AI tools across federal agencies, citing concerns about national security and alleged pressure on defense institutions. During the same period, the Department of Defense reportedly partnered with a competing AI firm, OpenAI, to deploy its technology in sensitive military applications.
The company has also faced internal data exposure incidents. Earlier in the year, confidential documents related to the unreleased model were unintentionally made accessible after being stored in a public data cache. In a separate case, more than 500,000 lines of proprietary code tied to its AI systems were accidentally published online. The firm described the incident as “human error, not a security breach.”
According to statements from a senior safety official cited in reports, the new model demonstrates an unprecedented level of independence and analytical capability. It was described as “extremely autonomous” and capable of functioning similarly to a highly skilled cybersecurity expert, including identifying vast numbers of system vulnerabilities and generating corresponding exploit methods—an ability not seen in earlier versions.
In additional remarks cited in reports, the same official emphasized the broader significance of the development, stating that it represents “the starting point for what we think will be an industry change point, or reckoning, with what needs to happen now.”
The company has indicated that access to the model will be tightly controlled, limiting its availability to a small group of technology and cybersecurity organizations. According to its official communications, a wider release will only be considered once sufficient safeguards are implemented to mitigate the risks associated with its advanced capabilities.
Legal Disclaimer:
MENAFN provides the
information “as is” without warranty of any kind. We do not accept
any responsibility or liability for the accuracy, content, images,
videos, licenses, completeness, legality, or reliability of the information
contained in this article. If you have any complaints or copyright
issues related to this article, kindly contact the provider above.
Most popular stories
Market Research
More Story
Comments
No comment