Search the evidence

Find the signal.

Search titles, impact clusters, countries, organizations and the full text of every analysis.

5 stories found

Work & marketsGlobal+1 clusters01

OECD, “Artificial Intelligence Markets: Recent Developments and Competition Issues”

The OECD finds a mixed competitive picture: foundation-model performance continues to improve while quality-adjusted prices decline and leadership changes hands, but structural concentration persists in the inputs that determine long-term market power, particularly advanced chips, cloud infrastructure, compute, proprietary data, and specialized talent. The report warns that vertical integration, first-mover advantages, and preferential partnerships between model developers and dominant chip or cloud providers could entrench a small group of firms even if the model layer currently appears dynamic.

2 min
PrivacyGlobal+1 clusters02

OECD Trust Survey chapter, “Trustworthy artificial intelligence in the public sector”

OECD’s 2026 trust-survey chapter finds that most people remain skeptical of AI deployment in the public sector, and that people are more likely to expect AI to improve service quality and efficiency than to expect it to be fair, transparent, privacy-protective, or subject to human oversight. This suggests that public-sector AI adoption is not just a technical capacity problem.

2 min
A mechanical clock and task trays with citrus enamel blocks on a petrol-blue work surface.
Work & marketsUnited States / Europe+1 clusters03

AI saves task time. Who gets the dividend?

The everyday case for AI at work deserves credit: people say it helps them get parts of their job done faster. Gallup's latest analysis finds that 52% of frequent AI users report spending less time gathering information and 47% less time writing. Those percentages describe people reporting a direction of change—not the share of working hours saved. The more interesting result is what does not move. Most frequent users report unchanged mental effort. Time spent checking and correcting work moves in both directions: 21% report more, 23% less and 56% no change. This is not evidence that every efficiency gain disappears into review. It is evidence that a faster task and a lighter job are different outcomes. An OECD/Eurofound paper helps frame that distinction, cautiously. Its historical analysis links occupational AI exposure with some changes in task composition, but finds little clear association with autonomy or teamwork. It largely covers the period before today's agents and measures exposure, not confirmed use. It cannot adjudicate whether a particular 2026 workplace deployment works. Our reading is encouraging but conditional. A useful assistant can remove drudgery; the organization decides whether the released capacity becomes better service, learning, shorter hours or simply another assignment. Workers and managers need evidence about the whole workflow, including quality and recovery from errors. Counting generated pages or faster first drafts leaves the distribution question unanswered. The next productivity story should tell us not only what became quicker, but who gained something worth keeping.

3 min
A rising AI investment tower feeds an autonomous shopping agent approaching a bank vault marked with identity, authorization, and liability gates.
Work & marketsGlobal+4 clusters04

AI capital props up growth as banks write voluntary rules for agents that spend

The OECD's outlook and a new banking-industry paper show AI entering the economy through two control points: investment and authorization. The OECD projects global growth of 2.9 percent in 2026 and 3.0 percent in 2027, with the United States at 2.2 and 2.1 percent, the euro area at 1.0 percent in both years, and China at 4.5 then 4.2 percent. It says AI investment has supported trade and activity, while warning that spending increasingly relies on external financing. If expected returns do not materialize, a correction could be amplified through lenders and markets. At the transaction layer, six banks have published principles for agentic commerce: transparency, safety, privacy and data, customer choice, and interoperability. They identify identity, authorization, fraud prevention, liability, and customer protection as necessary foundations when AI agents begin choosing and paying for goods. The principles are directional, not an implementation standard. A later paper will develop the blueprint. AI is already supporting macroeconomic demand while the rules for letting agents transact are still being written. A purchasing agent can create disputes about who authorized a payment, who bears fraud, and whether it optimized for the customer's interest. The next phase of AI risk may arrive not as a model failure in a lab, but as ordinary credit, payment, and liability exposure distributed through the financial system.

10 min
A classroom cutaway contrasts widespread chatbot access with a student and teacher checking an AI answer against evidence.
Cognition & learningOECD member and partner economies+2 clusters05

PISA finds AI access alone does not create a learning advantage

AI use in education is no longer a pilot program waiting for permission. PISA 2025 surveyed and tested more than 760,000 fifteen-year-olds across 91 countries and economies, and its OECD average shows 45.5% of students use AI at least weekly to help them learn. Yet the report does not find a simple more-use, more-learning relationship. After accounting for socio-economic background, weekly users performed similarly in science to non-users, while students reporting very frequent or occasional use tended to score lower. For summarising and preliminary research, moderate users outperformed both limited and frequent users, but non-users often still outperformed users overall. These are associations, not proof that AI caused the score differences. The sharper policy signal is about instruction. Roughly six in ten students said school lessons had asked them to assess AI-generated information, and students who combined frequent learning use with such opportunities showed a more promising pattern. Disadvantaged students were less likely to receive that practice. That turns the AI divide from a device question into a teaching question. Schools that merely provide chatbots may scale shortcut behavior, distraction, or shallow confidence. Schools that redesign assessment, teach source checking, and make students defend their reasoning may turn the same technology into a learning instrument. The next advantage will not belong to the students with the fastest answer. It will belong to those taught how to challenge it.

5 min