Why did not use Doge Grok?
Grok, the Musk’s AI model, seems not available for the DOGE mission because it was only available as a special model in January. Moving forward, Doge May rely on Grok, according to WIRED, where Microsoft has announced that it will start hosting Grok 3 models from Xai in AZURE AI this week, The Verge I mentionedWhich opens the models up to more uses.
In their message, the lawmakers urged their presence to investigate a conflict of his interest in his interest, with a warning of potential data violations and the announcement that Amnesty International, as Dog used, was not ready for the government.
“Without appropriate protection, feeding sensitive data in the artificial intelligence system puts it in the possession of the system operator – a tremendous violation of the confidence of public employees and employee and an increase in the risks of cybersecurity surrounding the data.” “Treat artificial intelligence models make errors frequently and show great biases-technology is simply not ready to use in making high-risk decisions without appropriate examination, transparency, censorship, and studies on the basis of handrails.”
Although the WIRED report appears to confirm that DOGE has not sent sensitive data from the “Rokka” emails to an external source, legislators want more scrutiny in artificial intelligence systems to deter “the risk of sharing personal or sensitive information in another way with the AI model employees.”
The apparent fear is that Musk may start using its own models more, and take advantage of government data that its competitors cannot access, with the possibility of these data being at risk of breach. They hope to force Duji to separate all artificial intelligence systems, but it seems to be vick Amnesty International for Federal Use “Agencies should remove barriers that prevent innovation and provide the best value for taxpayers.”
“While we support the federal government to integrate new and approved artificial intelligence technologies that can improve efficiency or effectiveness, we cannot sacrifice the security, privacy and appropriate use when interacting with federal data.” “We can also not overlook the use of artificial intelligence systems, which are often known for hallucinations and bias, in decisions related to ending federal workers or federal financing without transparency and sufficient monitoring of these models – the risk of talent loss and critical research due to defective technology or defective uses of this technology is very simply high.”
