⇧ [VIDÉO] You might also like this affiliate content (after the ad)
The rapid evolution of artificial intelligence (AI) has crossed a critical threshold, causing even experts to express concerns about the ability of generative tools to create content indistinguishable from reality. An anonymous source, an AI expert, shared with the website Axios that the latest developments in generative technologies, which are still not public, make it possible to produce images so realistic that it is sometimes impossible to distinguish them from reality. A development that is surprising in its speed.
According to this source, this technology could be widely accessible as early as 2024, in a context where social media platforms appear to be relaxing their policies against disinformation and reducing the number of teams responsible for its implementation, with notable consequences for ethical considerations and safety. . This trend is further exacerbated by massive layoffs in the departments involved.
However, this information should be treated with caution. For example, the intention of the anonymous source from Axios (an American news website) could be tinged with alarmism to promote AI technologies, or reflect excessive optimism about its possibilities.
However, these advances in generative AI come in a worrying context, with the approach of the controversial US presidential election and the recent use of AI in disinformation warfare during the conflict between Israel and Hamas. These developments highlight the importance of not underestimating the potential risks and implications of such technologies in sensitive contexts.
Struggling with regulations
In 2023, the rapid evolution of artificial intelligence (AI) image generators has caused great concern among experts. This year has been marked by continued warnings about the potential dangers of this technology. Even President Joe Biden expressed serious concerns after attending a screening of the film Impossible mission, an experience that shortly preceded the issuance of an executive directive by the White House. Although this directive is broad, it remains vague in the details.
Bruce Reed, assistant to the White House chief of staff, emphasized on the “PBS” channel that the president’s experience with the film had reinforced his concerns about the potential misuse of AI. While the reality is still far from the film’s doomsday predictions, the Biden administration and the US Congress have considered workarounds, such as adding watermarks to AI-generated videos, to help distinguish real from fake. However, this measure is considered ineffective by experts as it can be easily circumvented.
According to the anonymous source quoted by Axios, today’s AI challenges may seem small compared to the challenges we may face in the future. This year can therefore be seen as a pivotal moment when AI began to infiltrate various fields.
The political and security responses to these challenges are made complex by the elusive nature of generative technology and the speed of its evolution, which is often difficult for the general public to understand. This situation is exacerbated by the lack of attention to certain ethical policies in the field of AI. Initiatives such as strengthening filters, educating the public to recognize AI-generated content, and regulating and setting industry standards prove crucial to maintaining trust and integrity in tomorrow’s digital universe.