Hodlx Guest Post Send your message
Artificial intelligence (artificial intelligence) creates an ethical crisis of algorithm. By shining on this problem, we risk allowing governments and companies to control global conversation.
Both artificial intelligence technology and industry have gone equivalent. Censorship potential becomes larger every day.
Every year to two years since 2010, it has the mathematical force to train artificial intelligence systems more With a 10 factor, which makes the threat of control and control over public discourse more realistic than ever.
Companies all over the world Ranked Privacy and data governance as the best risk of Amnesty International, while censorship has not been registered on its radar.
Amnesty International Which can process millions of data points in seconds Censorship can be controlled by various means, including moderate content and information control.
LLMS (large language models) and content recommendations can filter information, suppression or mass participation.
In 2023, Freedom House highlighted that artificial intelligence enhances state -led censorship.
In China, CAC (electronic space management) has combined control strategy in obstetrics, which requires Chatbots to support “basic socialist values” and prevent the content that the Communist party wants to control.
Chinese artificial intelligence models, such as Deepseek’s R1, The censorship is already Topics such as the Tiananmen Square massacre, in order to publish state novels.
“To protect free and open internet, democratic policy makers Work along with civil society experts from all over the world Human rights -based standards must create for both governmental and non -governmental actors that develop or spread artificial intelligence tools. Conclude Freedom House.
In 2021, the University of California in San Diego found that artificial intelligence algorithms were trained in controlled data groups, such as the Chinese Bido Pike, which links the “democratic” keyword to “chaos”.
Models trained on sources not subject to control related to “democracy” with “stability”.
In 2023, the report “Freedom of Freedom on the Network” in Freedom House found that the global Internet freedom decreased for the thirteenth year in a row. A large part of the decline in artificial intelligence.
Twenty -two countries have laws that require social media companies to employ automatic systems for moderate content, which can be used to suppress discussion and demonstrations.
For example, JUNTA Military in Myanmar used Amnesty International to monitor telegram groups, detention of dissidents, and execute death sentences based on their participation. The same thing happened in Iran.
In addition, in Bearus and Nicaragua, governments sentenced individuals to Draconian prison for their online speech.
Freedom House Find At least 47 governments used comments to the influence of online talks towards their favorite novels.
And it found that last year, the new technology was used in at least 16 countries to plant the seeds of suspicion, the opponents of the touch, or affect the general debate.
At least 21 countries of digital platforms require use of machine learning to delete political, social and religious speech.
Reuters 2023 a report He warned that Deepfakes created from artificial intelligence and wrong information can “undermine the confidence of the public in democratic processes”, enabling the systems that seek to tighten control of information.
In the 2024 US presidential election, the images created from artificial intelligence showed Taylor Swift, Donald Trump, showed that artificial intelligence is already manipulating public opinion.
China provides an example of an artificial intelligence control.
Leakat data collection analysis Through Techcrunch in 2025, it revealed an advanced AI system designed to monitor topics such as pollution scandals, business conflicts and political issues in Taiwan.
Unlike the traditional keyword -based liquidation, this LLMS system is used to assess context and political science.
Researcher Xiao Qiang pointed out that such systems “greatly improved the efficiency and granules of control of the state -led information.”
The report of the Judicial Committee in the House of Representatives in 2024 was accused of NSF (National Science Corporation) of financing artificial intelligence tools to combat “wrong information” in Covid-19 and 2020 elections.
The report found that NSF funded the anthology control and advertising tools.
“In the name of controlling the alleged misleading information regarding Covid-19 and 2020 elections, NSF exports millions of dollars to university and non-profit research teams,” Read The report.
“The purpose of these projects funded by taxpayers is to develop control tools and propaganda tools that work in Amnesty International that governments and large technology can use to form public opinion by restricting views or promoting others.”
2025 wireless The report has been discovered Deepseek’s R1 model includes censorship candidates for both application and training levels, leading to blocks on sensitive topics.
In 2025, the Pew Research Center wiped Find 83 % of the United States adults were concerned about the wrong information driven by artificial intelligence, as many showed concerns about the effects of freedom of expression.
Pio conducted an interview with artificial intelligence experts, who said that artificial intelligence training can unintentionally enhance the current energy structures.
AI control processing
She called for a 2025 misplaced information to better reports to reduce fear -based invitations.
the The survey was found 38.8 % of Americans are somewhat anxious, 44.6 % are very anxious, about artificial intelligence role in spreading wrong information during the 2024 US presidential election, while 9.5 % did not occupy any concerns, and 7.1 % was not perceived for this issue completely.
The establishment of an open source ecosystem is AI’s utmost importance. This means that companies reveal the sources of the training and biases data group.
Governments must Create Amnesty International Organizational Frameworks to determine the priority of freedom of expression.
If we want a human future, instead of the AI’s technical technology defect, the artificial intelligence industry and consumers need to build courage to address control.
Manouk Termaater is a businessman and exporting Amnesty International, Founder and CEO of AI’s vertical studio. HE aims to make Amnesty International accessible to everyone. With a background in engineering and financing, it seeks to disable the artificial intelligence sector using accessible customization tools and computers at reasonable prices.
Follow us twitter Facebook cable
Disclosure: The views expressed in Daily Hodl are not an investment advice. Investors must do due care before making any high -risk investments in bitcoin, cryptocurrency, or digital assets. Please note that your transfers and trading on your own responsibility, and any losses you may bear are your responsibility. Daily Hodl does not recommend buying or selling any encrypted currencies or digital assets, and Hodl Daily Andersor is an investment. Please note that the daily Hodl participates in dependent marketing.
Image created: dalle3
