Knowledge
Reports
Reports by qasb on how AI changes work, technology and organizations, with sources and stated limits.
Latest research report · Topic report 10/2026
AI and work: signs of weaker entry opportunities
For young people and for hiring in AI-exposed occupations, mostly descriptive studies from Germany, France and Denmark find less favourable trends. October’s evidence shows no broad employment collapse.
All editions
- 10/2026AI and work: signs of weaker entry opportunitiesFor young people and for hiring in AI-exposed occupations, mostly descriptive studies from Germany, France and Denmark find less favourable trends. October’s evidence shows no broad employment collapse.Topic reportWorkOctober 27, 2026
- 10/2026AI and education: mixed evidence on learning with AIIn several trials, AI support improved performance while AI was available. A later advantage without AI is not reliably shown, and an AI tutor at a US university was associated with less participation.Topic reportEducationOctober 27, 2026
- 10/2026AI products: recurring limits in testingAlmost a fifth of US adults reported near-daily AI use in August 2026. In tests, however, AI models run into recurring limits: when evidence is missing, outside English and in long agent tasks.Topic reportProductsOctober 27, 2026
- 10/2026AI in organizations: more businesses use AIIn several countries more businesses used AI in 2026 than a year earlier, large firms far more often than small ones. Many users spend little or nothing on it, and evidence on effects rests mostly on self-reports.Topic reportOrganizationsOctober 27, 2026
- 10/2026AI in the public sector: benefits mostly estimated rather than measuredAudit reports and official transparency records show AI in justice, welfare and tax administration in the UK and the US. The benefits are mostly estimated by the agencies themselves and rarely measured.Topic reportPublic sectorOctober 27, 2026
- 10/2026AI in research: new results with caveatsStudies in mathematics, chemistry, physics and weather forecasting report new results with AI. Checks show limits: formalisations can diverge from the written proof, and agents often reproduce results only partially.Topic reportResearchOctober 27, 2026
- 10/2026AI and the economy: Trade in AI goods grows, productivity unclearOctober’s sources mainly measure trade, data centres and financing in the AI build-out, often in nominal values that include price effects. An economy-wide productivity effect has not been measured officially.Topic reportEconomyOctober 27, 2026
- 10/2026AI and society: growing but unevenly distributed useIn Germany, more people used AI tools in 2026 than in 2025, younger people far more often than older ones. In the US and Britain, few got election information from chatbots. Audits show risks in health and salary questions.Topic reportSocietyOctober 27, 2026
- 10/2026AI misuse and system security: many warnings, few real-world figuresOn AI misuse, the evidence is mainly warnings, individual cases and self-reports. In tests, agents exceeded their task, and injected content often redirected models, especially after being adapted to the target model.Topic reportAI securityOctober 27, 2026
- 10/2026Cryptography and quantum: weaknesses mainly in implementationsA break of the ML-KEM, ML-DSA and SLH-DSA standards is not evidenced. Weaknesses showed up in libraries, devices and legacy TLS configurations. Quantum computing made progress on building blocks, not on attacks.Topic reportCrypto and quantumOctober 27, 2026
Sample reports
They show the structure and presentation of a report with illustrative figures. They are not published findings and do not count as editions. They can be read in the app after signing in.
- 10/2026AI in cryptography: faster in code, bound by proofThe post-quantum migration creates years of review and conversion work. AI mainly takes on inventory, testing and documentation. Design, proof and sign-off remain a human responsibility.Sample reportCryptography and mathematicsOctober 2, 2026
Reports are editorial publications by qasb built from public sources. They contain no customer data and assess no individual person.