Report from CNBC
In Brief – The European Commission has announced a new Digital Services Act (DSA) investigation of X and its integration of AI chatbot Grok with functionality that allowed users to create and post realistic manipulated images that could include depictions of women, and sometime children, in revealing clothing and sexualized poses. The largest platforms, including X, face the strictest DSA obligations and are regulated directly by the European Commission. They will examine whether X diligently undertook to assess and mitigate risks related to the creation and dissemination of content by Grok that was illegal and would have negative effects in relation to gender-based violence, risks that the Commission said appear to have materialized. Following widespread complaints in markets globally that kicked up in early January, the company put in place safeguards to prevent Grok on X from editing images of real people into revealing clothing, although the capability was reportedly still useable on the stand-alone chatbot. The Commission also said that it is extending their ongoing DSA investigation into X’s recommendation algorithms that began in 2023.
Context – Some DSA critics have long argued that the law creates a framework for online censorship, but its creators have responded that it did not set online speech standards but only directed large platforms to establish and abide by standards and processes to assess and mitigate risks of objectionable content. No AI chatbots are currently regulated by the DSA, although some may be soon. It’s an open question how DSA regulators will address so-called AI “guardrails”. However, integrating a chatbot into a VLOP like X gives Commission digital regulators the chance to start an AI regulation process. And while AI guardrail regulation will be controversial, the most politically sensitive aspect of X’s DSA problems, at least from a US-EU bilateral perspective, remains the potential regulation of how the platform recommends controversial political content.