- The US Navy prohibited the use of New Chatbot Deepseek
- Deepseek is an AI Chinese detainee
- The chatbot has become a chatppt competitor
The new Chatbot of AI Deepseek recently caused agitation because of its market disturbance after its large open source model seemed seriously undervalued the existing models.
But Deepseek is a Chinese company, owned and operated by a hedge fund in Hangzhou, which has frightened American technological companies and government institutions, the US Navy asking all members to avoid using technology to “any capacity” , due to the “potential security and ethical concerns associated with the origin and the use of the model. »»
This decision would be part of the generative strategy of the AI of the Ministry of Information of the Ministry of Navy, and the recipients by e-mail were invited to “refrain from downloading, installing or using the Deepseek model ”.
AI confidentiality problems
Deepseek’s privacy policy would probably disrupt the conscious confidentiality of us, since the chatbot apparently collects the personal information of users, which is stored on servers in China.
However, it should be noted that it is not specific to Deepseek, and Chatgpt is also a nightmare of confidentiality. Most of us have probably used to affirmations of technological companies that collect our data, but that does not mean that we must forget that this happens – in particular with the names of the large and familiar industry.
But confidentiality policy is not the only concern, because Deepseek suffered from its success in the form of large-scale malicious attacks against the platform. The incident, most likely a distributed denial of service attack (DDOS), noted that the platform was forced to temporarily put a break from new registrations.
“Open Source models like Deepseek, while offering accessibility and innovation, are increasingly vulnerable to supply chain attacks triggered during large -scale cyberattacks,” said Adityya Sood, Vice -President of the IA security and strategy engineering in Aryaka.
“These attacks, where adversaries exploit dependence on third-party dependencies, pre-formed models or public benchmarks, can have serious consequences. Opponents can alter pre-formulated models by incorporating malicious code, deadlines or poisoned data, which can compromise downstream applications. »»
Via CNBC
