Business and technology journalists
Chatgpt manufacturer, Openi, complained that the rivals, including those in China, used their work to achieve rapid progress in developing their own artificial intelligence tools (AI).
Openi status – and other US companies – as world leaders in AI Dramatic is undermined this week Suddenly the appearance of Deepseek, a Chinese application that can imitate the performance of Chatgpt, obviously by a fraction of the cost.
Bloomberg reported that Microsoft explores whether the data belonging to the Openi – to which the main investor is used in an unauthorized way.
BBC contacted Microsoft and Deepseek for comment.
Openai concerns echoed the recently appointed white house “AI and Crypto Czar”, David Sacks.
Talking about Fox News, he suggested that Deepseek may have used the models that Openi developed to become better, a process called knowledge distillation.
“There is significant evidence that what Deepseek did here distilled the knowledge of Openai’s models,” Mr. Sacks said.
“I think one of the things you will see over the next few months is our leading AI company that takes steps to try to prevent distillation … It would definitely slow some of these Copycat models.”
In a statement, Openi said that Chinese and other companies were “constantly trying to distill models of leading American AI companies”.
“How we go ahead … It is critically important that we work closely with the US Government to best protect the most capable models,” he added.
‘Deceptive’ claims
Naomi Haefner, Assistant Professor of Technology Management at St. Gallen in Switzerland, she said that the issue of distillation could throw out the idea that Deepseek had created his product for a fraction of costs in doubt.
“It’s unclear if Deepseek really trained his models from scratch,” she said.
“Openai stated that they believe that Deepseek may have applied large amounts of data.
“If this is the case, then the claims about the training of the model are very cheap deceptive. Until someone repeats access to training, we will not surely know if such economical training is really possible.”
Crystal Van Ooster, Venture Partner AI at Opencean, agreed that “Deepseek made it clear on publicly available research by major American and European institutions and companies.”
However, it is not clear how problematic the idea of ”building” the work of others is.
This is especially true of AI, where the accusation of non -compliance with intellectual ownership rights was often leveled in the main American companies AI.
Security and ethics
US officials also consider implications on the national security of Deepseek’s origin, said the press secretary of the Karolina Leavitt White House.
“I talked with [the National Security Council] They study what [the national security implications] Maybe, “said Mrs. Leavitt, who also re -approved US President Donald Trump’s remarks the day earlier that Deepsek should be a awake call for the US technological industry.
The announcement comes after the US Navy allegedly banned its members to use Deepseek applications because of “potential security and ethical problems”.
According to the CNBC, the US Navy sent the E -Stimin to its staff, warning them not to use the Depseek application because of “potential safety and ethical worries associated with the origin and use of the model”.
The Navy did not immediately respond to the BBC News comment request.
Data security experts have warned users to be careful with tools, since it collects large amounts of personal data and stores them on servers in China.
Meanwhile, Deepseek says it was a target of cyber attacks. On Monday, he said he would temporarily restrict registrations because of “big malicious attacks” to his software.
The inscription displayed on the company website says registration could be occupied as a result of the attack.
Yuyuan Tantian, a channel of social media under the Chinese state -owned emitter of CCTV, claims that the company has faced “several” Cyber attacks in recent weeks, which have increased in “intensity”.
The American AI industry shook the apparent breakdown of Deepsek, especially because of the overcoming attitude that they were now far ahead of the race.
The murder of trade limitations was believed to have banned the Chinese approach to top chips.
Although China has encouraged investments in advanced technology to diversify its economy, Deepsek is not one of the great Chinese companies that developed AI models for rivalry with US Chatgpt.
Experts say that now they still have an advantage – home is some of the largest chips companies – and that it is unclear exactly how much Deepseek has built its model and how far it can go.
Additional reporting from Fan Wang
