UN panel lists pressing opportunities, risks of AI in preliminary report

MANILA, Philippines – The United Nations’ Independent International Scientific Panel on Artificial Intelligence released the first-of-its-kind preliminary report assessing the most pressing opportunities, risks, and impacts of AI.

The Panel, made up of 40 members and co-chaired by Yoshua Bengio from Canada and Maria Ressa from the Philippines, seeks to “present a balanced analysis” of AI’s risks and opportunities, and aims to provide policymakers with a shared base of knowledge and evidence needed to respond effectively and in a timely manner.

“The technology is transformative, but if the world keeps moving along this trajectory, humanity will fail to realize the gains it promises. The risks — to societies, to security, and to our species — are too high, and the forces driving AI forward are not the forces that will deliver its benefits,” said Ressa.

“AI capabilities are outpacing both scientific understanding and governments’ ability to adapt. With growing evidence of deceptive AI behavior, science currently cannot guarantee that as capabilities continue to increase, AI will not cause catastrophic harm, either on its own or due to malicious users. To act effectively, global policymakers must understand these systems. This Panel provides exactly that: a rigorous, shared scientific foundation to guide our collective way forward, ” added Bengio.

In the report’s executive summary, the Panel wrote, “Deployed and applied thoughtfully, AI can support progress towards achieving the Sustainable Development Goals, advance health science, and increase access to education.”

“At the same time, the rapid pace of technological development and the breadth of potential applications present policymakers with significant challenges. The rapid, unchecked deployment of the technology at scale also presents considerable risks, including harms to the mental health of users, potential use as a destructive tool, impacts on social, economic and environmental systems, and challenges associated with controlling the technology,” the Panel added.

The Panel’s importance

The Panel defines AI systems as “machine systems that, broadly speaking, perceive, learn and act. They infer from inputs how to generate outputs such as predictions, content, recommendations, actions or decisions, with varying degrees of autonomy and adaptiveness.”

The Panel said its analysis is in a distinct position for three reasons.

First, it proceeds from the premise that the UN is the foremost global forum on transboundary risks of a global scale.

Second, the Panel is a UN mechanism with a given mandate to scientifically assess the state, risks and capabilities of AI.

Lastly, the Panel states it has a scientific and not political mandate: “to document scientific evidence, consensus and disagreements, and which knowledge gaps remain urgent to address.” By providing governments and institutions the evidence base they need to act over time, the Panel’s recommendations remain policy-relevant but not policy-prescriptive.

“This scientific character should make its findings comparable across regions and resilient to political cycles,” the Panel added.

What the Panel sees and what gaps exists

Section two of the report deals with what the evidence shows for AI, and the Panel came up with the following statements relating to AI, supported by examples and further commentary:

• Artificial intelligence capabilities are advancing faster than the ability to measure or govern them
• Only a handful of actors have trained frontier artificial intelligence models
• The inputs and outcomes of artificial intelligence are geographically and linguistically uneven
• The artificial intelligence divide is not just about access, but about capacity to influence artificial intelligence development
• For artificial intelligence to be useful, it must be supported by an enabling environment
• Agentic artificial intelligence is a governance step change (in terms of opportunities and risks)
• Artificial intelligence can erode the shared reality
• Artificial intelligence is transforming human rights, including children’s rights

Section three, meanwhile, lists potential opportunities and risks within various domains of knowledge.

Generally, the Panel noted that AI “has shifted from passive pattern recognition towards active reasoning and autonomous action. The field is advancing rapidly from current reasoning models towards orchestrated agentic networks and, ultimately, self-improving systems.”

The Panel added that access to AI does not automatically create a benefit. Instead, it said, “Purpose-built, task-specific AI is delivering measurable, evidence-backed gains across science, health, education, and agriculture. These gains are real but conditional: they depend on local contextualization, adequate infrastructure, and human preparation.”

Further, AI can be used positively, but the gains from AI use are not automatic. In terms of economic benefit, the Panel said, “productivity benefits require complementary investment in skills, data and organizational redesign.”

It added, “the core unresolved question is distributional: who captures the surplus and what happens to labor, to developing economies, and to regulatory frameworks built for a different era across different industries?”

In terms of global security, the Panel also noted that “AI can enable harmful operations, become a target of attack and amplify existing threats.” These include cybersecurity threats, but the Panel also noted that environmental impacts are “growing significantly.”

The Panel also said that AI was “transforming human rights, democracy, and the information ecosystem through system-level changes that create both significant opportunities and structural risks to information integrity and civic participation.” If these risks are not addressed, the Panel said that society’s ability to reap benefits from AI will be undermined.

It added, unfortunately that there is already evidence of institutions using AI capabilities as a catalyst for or threat to human rights, such as freedom of expression, access to information, justice, health and development, and other human rights.

Section three ends by stating that policymakers have an evidence dilemma on their hands, where they could either make decisions on how AI should be governed with insufficient scientific grounding now or wait for evidence, which may come too late for any worthwhile intervention.

Meanwhile, section four of the report outlined the scientific gaps — where the evidence base on certain aspects of AI remain uneven or insufficient, leading the Panel to say it cannot yet draw confident scientific conclusions in these respects.

These include inconclusive evidence surrounding macroeconomic and productivity effects from AI adoption, the environmental aspects of AI adoption, the global AI supply chain, and effects on the individual and collective level, among other gaps.

The Panel added that the scope of the preliminary report did not address military applications of AI and lethal autonomous weapons systems.

A full copy of the report, alongside an executive summary, is available on this page. – Rappler.com