Are democratic societies ready for a future wherein AI algorithmically allocates limited ventilators or hospital beds during pandemics? Or one wherein AI fuels an arms race between the creation and detection of disinformation? Or do you influence court decisions with amicus briefs written to mimic the rhetorical and argumentative variety of Supreme Court justices?
Decades of research show that almost all democratic societies find it difficult to have nuanced debates about recent technologies. These discussions must incorporate not only one of the best available scientific evidence, but additionally the various ethical, regulatory and social points of its use. Difficult dilemmas posed by artificial intelligence are already occurring at a pace that’s overwhelming the flexibility of recent democracies to resolve these problems together.
Broad public engagement, or lack thereof, has long been a challenge in integrating recent technologies and is vital to overcoming the challenges involved.
Ready or not, unintended consequences
Finding a balance between the impressive possibilities of recent technologies equivalent to AI and the necessity for societies to think through each intended and unintended outcomes shouldn’t be a brand new challenge. Nearly 50 years ago, scientists and policymakers met in Pacific Grove, California, on the so-called Asilomar Conference to choose the long run of recombinant DNA research, or the transplantation of genes from one organism to a different. Public involvement and participation of their deliberations was minimal.
Societies are severely limited of their ability to anticipate and mitigate unintended consequences of rapidly evolving technologies equivalent to AI without good faith intervention from a broad range of public and expert stakeholders. And there are real downsides to limited participation. If Asilomar had sought such wide-ranging input 50 years ago, questions of cost and access would likely have been on the agenda, as would the science and ethics of using the technology. If that had happened, the dearth of affordability of recent CRISPR-based sickle cell anemia treatments, for instance, might have been avoided.
With AI, there may be a really real risk of making similar blind spots with regards to intended and unintended consequences which might be often not obvious to elites equivalent to technology leaders and policymakers. If societies fail to “ask the proper questions that folks care about,” science and technology scholar Sheila Jasanoff said in a 2021 interview, “then regardless of what the science says, you’ll , not producing the proper answers or options for society.”
Ethical debates needs to be at the guts of efforts to manage AI.
Even AI experts are concerned about how unprepared societies are to responsibly advance the technology. We examine the general public and political points of emerging science. In 2022, our research group on the University of Wisconsin-Madison surveyed nearly 2,200 researchers who had published on AI. Nine in ten (90.3%) predicted that there will probably be unintended consequences of AI applications, and three in 4 (75.9%) didn’t imagine that society is ready for the potential impacts of AI applications.
Who gets to have a say in AI?
Industry leaders, policymakers and scientists have been slow to adapt to the rapid spread of powerful AI technologies. In 2017, researchers and scientists met in Pacific Grove for one more small, experts-only meeting, this time to stipulate principles for future AI research. Senator Chuck Schumer plans to carry the primary in a series of AI Insight Forums on September 13, 2023 to assist policymakers within the Beltway take into consideration AI risks with technology leaders like Meta's Mark Zuckerberg and X's Elon Musk.
Meanwhile, there may be a hunger amongst the general public to assist shape our shared future. Only a few quarter of U.S. adults in our 2020 AI survey agreed that scientists should have the opportunity to “conduct their research without consulting the general public” (27.8%). Two-thirds (64.6%) believed that “the general public must have a say in the applying of scientific research and technology in society.”
The public's desire for participation is coupled with a widespread lack of trust in government and industry to shape AI development. In a 2020 national survey conducted by our team, fewer than one in 10 Americans said they “mostly” or “very” trust Congress (8.5%) or Facebook (9.5%) to guard the well-being of the world Keeps society in mind when developing AI.
Algorithmic bias is just one problem related to artificial intelligence.
A healthy dose of skepticism?
The public’s deep distrust of key regulatory and industry players shouldn’t be entirely unwarranted. Industry leaders are finding it difficult to separate their business interests from efforts to develop an efficient regulatory system for AI. This has led to a fundamentally chaotic political environment.
It's not all the time problematic when tech corporations help regulators think through the potential and complexity of technologies like AI, especially once they disclose potential conflicts of interest. However, the contribution of technology leaders to technical questions on what AI can or might be used for is just a small a part of the regulatory puzzle.
More urgently, societies must work out what kinds of applications AI needs to be used for and the way. Answers to those questions can only emerge from public debates that engage a wide selection of stakeholders on values, ethics and fairness. Meanwhile, public concern concerning the use of AI is growing.
AI may not wipe out humanity anytime soon, but it surely is more likely to turn out to be increasingly disruptive to life as we currently comprehend it. Societies have a limited window of opportunity to search out ways to have interaction in debates in good faith and work together toward meaningful AI regulation to make sure that these challenges don’t overwhelm them.
