Prohibit the input of unpublished text content. The United States will require employees to only use the paid version of ChatGPT Congress | Paid | Employees
The US House of Representatives only allows members to use the paid version of ChatGPT Plus.
On June 26th local time, news website Axios reported that the US House of Representatives has set new regulations on how Congressional offices use ChatGPT, allowing only employees to use paid versions of ChatGPT Plus, and non ChatGPT chatbots have not yet been licensed for use.
ChatGPT Plus is a paid version of the chatbot ChatGPT, priced at $20 per month. Through Plus, users can access the company's multimodal model GPT-4 released in March this year, as well as the previously released GPT-3.5.
In a memo, the Chief Executive of the House of Representatives, Catherine L. Szpindor, stated that lawmakers and staff can only use the paid version of ChatGPT Plus because the Plus version provides "important privacy protection features such as congressional data.". In addition, the office can only use the product for "research and evaluation" when privacy settings are enabled, and it is prohibited to paste "any unpublished text" into the model.
Szpindor added in the memo that the House has not authorized the use of other versions of the ChatGPT and other large language models.
Axios stated that this is the latest example of the US government's response to the explosive growth of generative artificial intelligence at both the legislative and individual levels.
Some media believe that this policy in the House of Representatives will not face much opposition, as both parties in Congress are trying to regulate and govern AI. In the House of Representatives, Representative Ritchie Torres proposed a bill requiring a disclaimer for the use of generative AI, while another Representative, Yvette Clark, expressed a desire for similar disclosure of political advertisements.
On the Senate side, a few days ago its majority leader Chuck Schumer called on Congress to pass new legislation as soon as possible to regulate the AI industry. He proposed a new framework that detailed the key areas that Congress should focus on, including the potential risks of AI in national security and job hunting leading to unemployment.
Schumer stated last week that "AI may be the most eye-catching innovation to date, as it is a force that can ignite new eras such as technological progress, science, and industry." At the same time, he emphasized that "the first issue we must address is to encourage innovation, not stifle it.".
In addition to the United States, the European Union is also making steady progress in AI regulation. On June 14th, the European Parliament voted to pass the Artificial Intelligence Act, which prohibits real-time facial recognition and imposes new transparency requirements on generative artificial intelligence tools such as ChatGPT.
After the bill is passed, in extreme cases, violating companies may be fined up to 30 million euros, or 6% of the company's global annual revenue, which means that for large technology companies such as Google and Microsoft, their fines may reach billions of euros.
In addition to government agencies, generative AI such as ChatGPT has also faced restrictions from private companies such as Samsung and Apple - they are concerned that confidential data may be leaked through these tools, as OpenAI has indeed exposed user chat records before.