Republished with full copyright permissions from The San Francisco Press.

There’s been a recent surge of discussions centering around the potential implications of the bill introduced by Senators Josh Hawley and Richard Blumenthal, to explicitly exempt AI from Section 230. This move to “hotline” the bill, in an attempt to expedite its passage, has sparked concerns within the tech and legal communities.

At the heart of the matter lies the fundamental question of whether the output of generative AI systems should fall under the protection of Section 230. While some argue against it, citing that the results are directly attributed to the company in question and not third-party speech, legal experts have emphasized that courts have already recognized the protection of algorithmic output derived from content provided by others. This existing legal precedent is crucial in safeguarding algorithmically generated content such as search snippets and autocomplete suggestions.

The bill’s broad and ambiguous definition of “generative artificial intelligence” raises significant concerns regarding its potential implications. By encompassing technologies that extend beyond what is commonly understood as AI, including tools like autocomplete and spellcheckers, the bill’s scope appears to be overly expansive. Furthermore, its exemption from Section 230 extends not only to the output of generative AI but also to any conduct involving the use or provision of such technology. This fails to establish a clear and reasonable limitation, given the pervasive integration of AI in various digital products and services.

One of the most alarming aspects of this bill is its potential to erode the protections afforded by Section 230 from a wide array of internet companies. With the ever-growing incorporation of AI across digital platforms, the bill’s far-reaching implications could render these companies susceptible to legal liabilities, even for user-generated content that incidentally involves AI elements. Such undue exposure to legal risks could stifle innovation and creativity in the online space, deterring companies from offering AI-driven tools and services due to fear of heightened liabilities.

Moreover, the bill’s carve-out of state law from Section 230 preemption could open the floodgates to a patchwork of conflicting state regulations, further complicating the legal landscape for internet companies. This loophole poses the risk of incentivizing states to enact laws that introduce liability for AI-related content, thereby fragmenting legal standards and undermining the overarching protections established by Section 230.

In essence, the proposed bill represents a misguided legislative approach that jeopardizes the foundational principles of the open internet. Its implications extend beyond AI itself, posing a direct threat to the innovation and vibrancy of digital platforms. By subjecting internet companies to sweeping liabilities and creating a climate of legal uncertainty, the bill undermines the very essence of Section 230, a cornerstone that has propelled the United States to the forefront of the global internet economy.

As we navigate the evolving landscape of generative AI, it is imperative to approach legislative reforms with precision and foresight. Rather than resorting to ill-conceived measures that could compromise the integrity of the internet, policymakers should seek nuanced solutions that balance technological advancements with legal safeguards. The future of the internet’s vitality hinges on the preservation of a balanced and effective legal framework, one that fosters innovation while upholding fundamental principles of free speech and digital freedom.