AI's Arrival Coincides With Erosion of Trust, Raising Concerns
The development of sophisticated AI technologies is occurring at a time when societal trust in institutions and information sources has significantly declined.
Artificial intelligence is emerging in a world ill-equipped to handle its power due to a systematic dismantling of foundational trust over the past two decades, according to analysis from Time. The current era is characterized by a "post-trust" landscape, where shared mechanisms for establishing facts have broken down, making the responsible deployment of advanced AI technologies increasingly challenging.
This decline in trust is evident across various sectors. Over the last twenty years, the information ecosystem has shifted from curated news delivered by established intermediaries to a fragmented digital space where anyone can publish, leading to the amplification of misinformation. Algorithms optimizing for engagement over accuracy have further exacerbated this trend. Consequently, a significant majority of people now rely on algorithmic systems for their news, outsourcing their information consumption to platforms they do not fully understand or trust.
In the United States, public trust in the federal government has fallen from over 70% in the late 1950s to under 20% today. Confidence in mass media has also diminished, with fewer than a third of Americans expressing a fair amount of trust. This erosion extends to other institutions, including religious bodies, financial systems, healthcare organizations, and even science itself.
The "post-truth" era, where emotional appeals often overshadow objective facts, has deepened into a "post-trust" era, marked by a loss of shared procedures for determining reality. The advent of AI-generated content, such as convincing images of public figures, highlights this vulnerability. What was once a world where seeing was believing has become one where fabricated evidence can be produced at scale, further complicating the challenge of discerning truth.
Research indicates that exposure to opposing political views on social media can increase polarization, and false news spreads significantly faster and wider online than trustworthy news. Furthermore, cognitive research suggests that people process information that confirms their beliefs differently from information that challenges them, contributing to deep societal divisions.
Into this fractured environment arrives AI, a technology with cognition as its core competence. Unlike previous technologies that amplified or executed human intent, AI generates arguments, synthesizes information, and produces creative work that mimics human thought. These models learn by identifying patterns in vast datasets, which include not only knowledge but also the distortions, biases, and conflicts present in human-generated data.
The result is that AI reflects a refracted version of humanity, trained on data from a post-trust world. It provides outputs that are difficult to audit, based on opaque training data and optimized for objectives that are not always controllable. This new form of cognitive infrastructure is emerging precisely when the social infrastructure for trust is under severe strain, posing significant challenges for society.