Yomiuri, NTT release joint proposal on use of generative AI

The Yomiuri Shimbun and Nippon Telegraph and Telephone Corp. on April 8 released the “Joint Proposal on Shaping Generative AI” in a bid to ensure discussion spaces in which human liberty and dignity are maintained amid the rapid expansion in the use of AI. The proposal mainly seeks to balance the control and use of generative AI from both institutional and technological perspectives, with a call for legislation to restrict the use of such technology in elections and security.

The advantages of generative AI are that it can be used by anyone via the internet and that it can be expected to improve labor productivity to a certain degree, according to the proposal. On the other hand, the proposal also pointed out that generative AI is a technology that currently cannot be fully controlled by human beings and could lead to situations in which generative AI “lies with confidence” and people are “easily fooled.” The proposal warns there is a risk that journalism and academic research that have provided accurate and valuable information may collapse under such circumstances.

In this regard, the proposal emphasized, if the technology is not properly controlled, “in the worst-case scenario, democracy and social order could collapse, resulting in wars,” thus “measures must be realized to balance the control and use of generative AI from both technological and institutional perspectives, and to make the technology a suitable tool for society.”

On the technical side, the proposal highlighted the effectiveness of “Originator Profile (OP),” a technology that identifies the senders of information online so that users can confirm the reliability of information.

On the institutional side, elections and security are cited as risk areas that require special caution. The proposal stated that “the unconditional application of [generative AI] … could cause enormous and irreversible damage,” and that hard laws “should be introduced without hesitation.” It also referred to how intellectual property rights, such as copyrights, should be properly protected.

Regarding rulemaking, including legislation, the proposal acknowledged that “it is expected to require a long time and involve many obstacles to develop” a “strategic and systematic data policy.” Taking this into consideration, it calls for addressing the “strategic and systematic data policy” as a long-term issue and working on “individual regulations” and “effective measures” as shortterm issues.

As an issue that urgently needs to be addressed, the proposal highlights the problem of the adverse effects of “the attention economy,” a method of attracting attention with information of radical and extreme content, being amplified by generative AI. The proposal described the problem as “the out-of-control relationship between AI and the attention economy,” and stressed with a strong sense of urgency that it “has already damaged autonomy and dignity, which are essential values that allow individuals in our society to be free. These values must be restored quickly.”

To ensure discussion spaces in which human liberty and dignity are maintained, it is necessary to ensure that there are multiple AIs of various kinds and of equal rank that “users can refer to them autonomously, so that they do not have to depend on a specific AI.”

The Yomiuri Shimbun and NTT Corp. in cooperation with Keio University's Cyber Civilization Research Center

have since last fall studied with experts to compile the proposal. These parties will

continue to study and make proposals on generative AI in the future.