Pentagon is embracing Musk’s Grok AI chatbot because it attracts international outcry : NPR

[ad_1] <img src="https://npr.brightspotcdn.com/dims3/default/strip/false/crop/5600x3733+0+0/resize/1100/quality/50/format/jpeg/?url=http%3A%2F%2Fnpr-brightspot.s3.amazonaws.com%2Fdc%2F3f%2Ff005bf0544a38e1716d92968a16c%2Fap26007557270287.jpg" data-template="https://npr.brightspotcdn.com/dims3/default/strip/false/crop/5600x3733+0+0/resize/{width}/quality/{quality}/format/{format}/?url=http%3A%2F%2Fnpr-brightspot.s3.amazonaws.com%2Fdc%2F3f%2Ff005bf0544a38e1716d92968a16c%2Fap26007557270287.jpg" class="img" alt="Defense Secretary Pete Hegseth arrives for a classified briefing with senators on the situation in Venezuela, at the Capitol in Washington, Wednesday, Jan. 7, 2026." fetchpriority="high"/> Protection Secretary Pete Hegseth arrives for a labeled briefing with senators on the state of affairs in Venezuela, on the Capitol in Washington, Wednesday, Jan. 7, 2026. J. Scott Applewhite/AP conceal caption

toggle caption

J. Scott Applewhite/AP WASHINGTON — Protection Secretary Pete Hegseth stated Monday that Elon Musk's synthetic intelligence chatbot Grok will be a part of Google's generative AI engine in working contained in the Pentagon community, as a part of a broader push to feed as a lot of the navy's knowledge as attainable into the growing expertise.

"Very quickly we could have the world's main AI fashions on each unclassified and labeled community all through our division," Hegseth stated in a speech at Musk's area flight firm, SpaceX, in South Texas.

The announcement comes simply days after Grok — which is embedded into X, the social media community owned by Musk — drew international outcry and scrutiny for producing extremely sexualized deepfake pictures of individuals with out their consent.

Malaysia and Indonesia have blocked Grok, whereas the U.Ok.'s unbiased on-line security watchdog introduced an investigation Monday. Grok has restricted picture era and enhancing to paying customers.

Hegseth stated Grok will go reside contained in the Protection Division later this month and introduced that he would "make all acceptable knowledge" from the navy's IT programs accessible for "AI exploitation." He additionally stated knowledge from intelligence databases can be fed into AI programs.

Hegseth's aggressive push to embrace the still-developing expertise stands in distinction to the Biden administration, which, whereas pushing federal companies to provide you with insurance policies and makes use of for AI, was additionally cautious of misuse. Officers stated guidelines had been wanted to make sure that the expertise, which might be harnessed for mass surveillance, cyberattacks and even deadly autonomous gadgets, was getting used responsibly.

The Biden administration enacted a framework in late 2024 that directed nationwide safety companies to develop their use of probably the most superior AI programs however prohibited sure makes use of, resembling purposes that will violate constitutionally protected civil rights or any system that will automate the deployment of nuclear weapons. It's unclear if these prohibitions are nonetheless in place underneath the Trump administration.

Throughout his speech, Hegseth spoke of the necessity to streamline and pace up technological improvements inside the navy, saying, "We'd like innovation to return from anyplace and evolve with pace and objective."

He famous that the Pentagon possesses "combat-proven operational knowledge from 20 years of navy and intelligence operations."

"AI is barely pretty much as good as the information that it receives, and we will be sure that it is there," Hegseth stated.

The protection secretary stated he needs AI programs inside the Pentagon to be accountable, although he went on to say he was shrugging off any AI fashions "that will not let you combat wars."

Hegseth stated his imaginative and prescient for navy AI programs implies that they function "with out ideological constraints that restrict lawful navy purposes," earlier than including that the Pentagon's "AI won't be woke."

Musk developed and pitched Grok as a substitute for what he known as "woke AI" interactions from rival chatbots like Google's Gemini or OpenAI's ChatGPT. In July, Grok additionally precipitated controversy after it appeared to make antisemitic feedback that praised Adolf Hitler and shared a number of antisemitic posts.