
What to know
- Check how hiring is changing in your field as well as how many people are leaving it.
- When pricing a project, account for the hours needed to review and deliver the work.
The week in three questions
This briefing covers November 9–15 through 9:00 a.m. Singapore time on November 15. This week’s sources ask three different questions: how quickly people find work, how clients put a price on assignments, and how people entering a profession can still learn the basics as AI takes on routine tasks.
Finding a job after unemployment
Cleveland Fed researchers tracked US adults aged 23–54 in linked monthly survey records. Comparing January 2024–April 2026 with 2015–2019, they found a roughly six percentage point larger fitted decline in job finding at the highest versus lowest task exposure to AI. The relationship was weaker using Claude usage as a proxy. The separation results showed no consistent pattern, and unemployed people without a prior occupation were outside the comparison.
Australian central bank staff found little evidence of a sustained gap in quarterly job-finding rates by exposure through early 2026. They did see tentative signs of weaker employment and hours among younger people in more exposed occupations. The studies use different exposure rankings and transition measures, and neither isolates the effect of AI.
For separate US labor-market context, initial unemployment insurance claims rose by 9,000 to a preliminary 200,000 in the week ending November 7. The seasonally adjusted four-week average stayed at 195,250. Claims count applications for benefits and provide no breakdown by AI exposure.
Sources: Federal Reserve Bank of Cleveland; Reserve Bank of Australia; US Department of Labor
Freelance budgets and the time behind them
OpenAI and Upwork compared listings through April 2026. After adjustment for work subcategory and posting month, AI-related listings had about 52% higher posted fixed-price budgets and 1.5% higher posted hourly rate midpoints. Among hired hourly contracts with positive recorded hours, AI-related work logged about 41% more hours in the first 90 days using the same adjustment. The comparisons describe separate groups on one marketplace. Higher advertised budgets do not establish higher realized earnings.
Sources: OpenAI and Upwork
Training to check the work
The Institute of Singapore Chartered Accountants signed a five-year agreement with Singapore’s five polytechnics and the Institute of Technical Education. Plans include industry projects, mentoring and internships. ISCA reported more than 3,000 polytechnic and ITE students had enrolled in its existing AI Fluency Programme and supplied no completion or placement figures.
In Middlesex County, New Jersey, Rutgers researchers held six focus groups with 21 small-business owners in May–June 2026. Participants described reviewing AI outputs and wanting practical help suited to their industries. The voluntary group was small and already experimenting with AI. Its accounts cannot establish how widely businesses use the tools or what happens to staffing.
Sources: Institute of Singapore Chartered Accountants via Media OutReach Newswire; New Jersey State Policy Lab and Heldrich Center
Why adoption and productivity can move at different speeds
In a November 13 speech, Federal Reserve Vice Chair Philip Jefferson described evidence of efficiency gains in specific tasks and mixed evidence on broader productivity. He pointed to limited use across business functions and possible costs elsewhere in a workflow. This was his interpretation of existing studies and data.
Sources: Federal Reserve Board
Our interpretation: show how you checked your work
Pick a task from a real opening or freelance brief. Make a small sample using public or fictional data. Record the time you spend producing a draft, checking it and correcting any errors. Put one correction in your portfolio and explain the evidence that led to it. If you quote a price or assess a training course, keep that review time in the estimate.
Sources & method
AI-generated weekly synthesis of primary studies, an official release, an issuer announcement and a policy speech. The portfolio exercise is editorial interpretation.
- Job Finding and Unemployment in the Age of AI Federal Reserve Bank of Cleveland ·
- Monitoring Early Effects of AI on the Labour Market Reserve Bank of Australia ·
- Unemployment Insurance Weekly Claims US Department of Labor ·
- The Rebundling of Freelance Work in the AI Age OpenAI and Upwork ·
- ISCA announces its partnership with the polytechnics and ITE Institute of Singapore Chartered Accountants via Media OutReach Newswire ·
- Artificial Intelligence and Small Businesses in New Jersey New Jersey State Policy Lab and Heldrich Center ·
- New Technology, Old Questions: Productivity Puzzles and the Macroeconomy Federal Reserve Board ·
Drafted and source-checked by Codex AI agents. Read our sourcing, illustration, and correction policy.
Explore the occupations
What could this mean for your work?
See task exposure and potential career moves for roles connected to this story.
