The AI industry remains in its early stages, and some experts argue that waiting a few more years could result in more informed and effective regulatory decisions. “The AI industry is very nascent and we would probably be able to make better choices if we waited five years,” one policymaker noted.
One area of contention involves a new set of voluntary guidelines aimed at developers of general-purpose AI models, including major players like OpenAI’s GPT and Google’s Gemini. While intended to ensure safety and accountability, these proposals have drawn criticism from tech companies concerned about the added regulatory burden.
However, civil society organizations strongly oppose any moves to ease existing or proposed regulations. Their resistance is expected to keep discussions around revisions to AI legislation as contentious as when the original rules were drafted.
In fact, the European Commission’s latest AI strategy report — published Wednesday — reflected ongoing debates behind the scenes. A key section on simplification was significantly toned down from a leaked earlier draft. Originally referencing an “opportunity to minimize the potential compliance burden,” the final document softened the language, instead calling for efforts to “facilitate compliance with the AI Act.”
Additionally, two prominent EU lawmakers criticized the Commission’s decision to drop a proposal for a bloc-wide liability scheme that would determine responsibility for harm caused by AI. They described the decision as “premature and unconvincing.”
Maximilian Gahntz, AI policy lead at Mozilla, cautioned that the pursuit of simpler rules should not compromise the core objectives of the EU’s AI Act. “Simplification should not mean deregulation,” he said. Gahntz emphasized that any streamlining efforts must not come at the cost of safety, accountability, or the law’s intended protections.
Leave a Reply