Trump says, “Amnesty International” in Washington says, and he launches the artificial intelligence action plan

US President Donald Trump has pledged to preserve the “Amnesty International” models from Washington and turn the country to the “Amnesty International Export Force” by signing three executive orders focusing on artificial intelligence on Wednesday.

Exit from Diversity, fairness and integration initiatives (DEI) – Paratory term that includes many practices, policies and strategies aimed at promoting a more comprehensive and fair culture – was a great axis in the second Trump administration. Now, the White House brings the battle to artificial intelligence.

the “Preventing artificial intelligence in the federal governmentIt stipulates that the federal government “is committed not to buy models that sacrifice honestly and accurately on ideological business schedules.”

Dei is defined as one of the “most prevalent and destroyed” of these ideologies that will be removed from the artificial intelligence models used by the government.

“LLMS should be neutral and non -partisan tools that do not manipulate responses in favor of ideological beliefs such as Dei,” Llms said, adding that developers should not deliberately encrypt partisan or ideological rulings in LLM outputs unless users are called.

As he admitted in order, the use of artificial intelligence is increasingly spread across the Americans daily, and is expected to play an important role in the way they learn and consume information. This can give the outputs of artificial intelligence models of importance and influence at the community level.

In the eyes of the Trump administration, Dei can lead to artificial intelligence to discriminatory results; Distorting and treating the outputs of the artificial intelligence model with regard to race and sex; Including concepts such as critical race theory, sexual bias, unconscious prejudice, and systemic racism.

“Dei replaces adherence to the truth in favor of preferred results, and as modern history shows, an existential threat to AI is reliable,” says it.

Without giving details, arrangement indicates previous examples, including the main Amnesty International model that changed the race or gender of historical characters such as the Pope and the founding fathers when demanding pictures.

In response to the reaction last year, Google He has withdrawn The Gemini Ai images feature, saying it has offered “inaccuracy” in historical photos. After months, the company was launched to update Issue.

Instead of “wakes up Amnesty International”, the government must buy “models of artificial intelligence” that overlook the truth, “gives priority to historical accuracy, scientific and objective inquiries, and recognizes the uncertainty when reliable information is incomplete or contradictory,” and stated.

However, it adds that the federal government “must be hesitant” in organizing the functions of artificial intelligence models in the private market.

In the developments of other Amnesty International on Wednesday, the Trump administration I signed an order To stimulate innovation in technology by removing what he called “arduous federal regulations that hinder the development and dissemination of artificial intelligence.”

Another matter aims to create and implement “American artificial intelligence exports programTo support the development and deployment of AI technology in the United States abroad.

The movements are part of the administration “Winning the Artificial Intelligence Racing: The Plan of Artificial Intelligence in America“It says it defines 90 federal political procedures across three columns: accelerate innovation, building the Amnesty International infrastructure, and leadership in diplomacy and international security.

Executive requests come Amnesty International after the receipt of artificial intelligence companies Government contracts With the Ministry of Defense, it gives them up to $ 200 million to help accelerate the agency’s adoption of advanced artificial intelligence capabilities to “address critical national security challenges.”

The formulation of anti -artificial intelligence appears to be in line with a lot of Xai messages, run by former Trump advisor and Megadonor Elusk. The company has been announced by the company as a “fight of tourists” and “Searching for the truth to the maximum extent“artificial intelligence.

However, in the last weeks, a large number of major headlines surrounding GROK have highlighted the differences that could arise regarding how to train artificial intelligence models and their perceived political biases.

When a new version of Grok 4 was launched earlier this month, CNBC discovered that she would seek MUSK opinions while answering some controversial questions. xi He later said In a post that this was a mistake and that he had amended the problem.

In the same post, the company taught that GROK 4 has temporarily indicated itself as “mecahitler” in response to some users. This came a few days after the company apologized for the free Chatbot Grok 3 that praised the Adolf Hitler and gave other anti -Semitic and controversial comments on the social media platform X.

The Grok ordeal happened after spending Musk agreed The user’s claim that LLM from Xai was affected by the left bias. In early July, he announced that his team had made changes and “improve” Chatbot.

The Hitler Grok episode sparked a wide and increased condemnation Checking the European Union.

Countries all over the world are still their organizational guidelines for the Chatbots AI line. In the United States, there is no comprehensive federal legislation or regulations that regulate the development of artificial intelligence, specifically prohibited or restricted its use. a report this week.