Mixed learning groups broaden teachers’ AI use
With identical training, teachers in mixed-specialty groups report AI use that is 0.19 standard deviations broader. The study measures self-report, not quality.
AI RADAR · 18 STUDIES
All studies in the AI Radar
A new study every day, put in context for companies. Newest first. Back to the AI Radar →
AI narrows the performance gap between education groups
On a business case, AI narrows the education gap from 0.548 to 0.139 standard deviations. Durable learning remains untested.
AI improves performance on more demanding IT tasks
About 65 versus 39 percent of available points on architecture tasks. Control group without internet; durable learning not measured.
More speed is not deep transformation
81 percent report more speed, 25 percent deep transformation. Separate questions, not measured productivity.
AI-supported work needs checking before handoff
The study captures expectations of colleagues. It does not show how often AI results are actually checked.
More planning improves AI writing but takes time
Goal-setting and task-breakdown support improved writing quality. Participants nevertheless rated the AI as less helpful.
AI creates new tasks for directing agents
An open-ended survey identifies new work directing agents, setting standards and building AI systems.
AI agents need instructions and checking
Researchers observed work with coding agents. This time share describes prompting and checking, not wasted time or a measured productivity gain.
AI incidents reach the top of the organisation
A reported incident may involve a wrong AI decision or exposed data. The survey captures escalations, not control effectiveness.
More AI contributions need enough review time
The survey captures experiences in open-source projects. More contributions demonstrate neither higher quality nor measurable business value.
83% report no productivity impact from AI yet
No reported impact does not mean AI is ineffective. The survey captures perceptions, not independently measured changes.
Using AI does not mean embedding it in core operations
82% of AI-using program alumni say AI is not yet central to their business. Self-report, not a general small-business rate.
More speed with AI does not remove every bottleneck
AI-related effort also includes reviewing results and correcting prompts.
Using AI does not yet mean embedding it in daily work
The Scottish survey distinguishes individual AI applications from embedding AI across most of an organization. It measures reported use, not productivity or business performance.
At Anthropic, more Claude sessions focus on instructions and oversight
The average session share for instructions, documentation and oversight rises from 10% to 22%. This measures AI use, not working time.
Few respondents report rules for stopping AI initiatives
31% report rules before production; 17% for AI already in use. Two separate practices, based on self-reports.
Organised team chats: more correct AI answers in the test
The study tested how an AI uses information from internal team chats. It found better answers, but does not establish time savings in everyday work.
AI can fail again on the next attempt
20 of 50 tasks passed all three attempts with GPT-6 Sol. One success does not establish repeatability.