Welcome to this week’s StaffingHub Brief, your strategic intelligence roundup for staffing agency leaders. In this week’s issue:
- The SIA Staffing Confidence Index hit 128.0 in July, near its post-pandemic peak, while US staffing hours rose 10% year-over-year at the fastest pace since 2022.
- A Federal Reserve study of 490,000 earnings calls found 95% of AI productivity claims reference future gains; actual productivity growth across public companies sits at 0.07%.
- Two AI notetaker lawsuits filed in late July expose employers to $5,000-per-violation liability in California, and the EU AI Act classified all hiring AI as high-risk as of August 2.
Confidence and volume are both moving
The SIA Staffing Confidence Index came in at 128.0 in July, just below May’s 128.8 reading, the highest since 2023. New orders are the standout: a net 62% of firms reported increases over the past three months, the best reading since late 2021. The gap between current conditions (124.7) and six-month expectations (131.3) is narrow, which SIA reads as firms viewing recent gains as sustainable rather than temporary. (Learn more)
US staffing hours rose 10% year-over-year for the week ended July 25, the fastest growth rate since June 2022, excluding holiday weeks. Industrial led at 15% YoY, driven by manufacturing, logistics, and data center demand. Commercial hit 12% YoY. Office and clerical was the one outlier, down 5%. (Learn more)
Why it matters: The sentiment-to-volume gap that defined 2024 is closing. Both are moving at the same time, which gives the confidence reading something real behind it.
The big firms are posting real revenue growth, and manufacturing is driving it
TrueBlue posted Q2 revenue of $443 million, up 11.8% year-over-year and above its own 2–8% forecast. PeopleReady led with 23% growth on energy sector strength and on-demand recovery. (Learn more) Adecco’s Q2 came in at €5.99 billion, up 5.6% organically. Fifth consecutive quarter of growth, with 160 basis points of market share gains and EBITA up 21%. (Learn more)
Manufacturing is the underlying driver. The ISM Manufacturing PMI rose to 55.6% in July, the fastest expansion in over four years. The employment subindex hit 52.8%, expanding for the first time in nearly three years. Fifteen of 16 manufacturing industries reported growth; transportation equipment, food and beverage, and computer and electronics led hiring. (Learn more)
Why it matters: Industrial and manufacturing demand is now showing up in hours billed, quarterly revenue, and sector employment data at the same time. Industrial is where demand is concentrated right now.
AI productivity is 95% promise, 5% reality
The St. Louis Fed studied 490,000 earnings call transcripts from 5,198 publicly traded US companies and found that 95% of AI-related productivity sentences reference future gains, compared to 75% for non-AI productivity mentions. Actual utilization-adjusted total factor productivity growth over the four quarters ending Q1 2026: 0.07%. Executives are describing current AI investments as future-state benefits in their own investor calls. (Learn more)
Adecco, meanwhile, has announced it will run 50% of its global revenue through agentic AI by December 31, backed by an unlimited Salesforce Agentforce license across 27,000 recruiters in 60-plus countries. The UK operation demonstrated 15% time savings, higher fill rates, and lower cost-to-serve before the global target was set. Only 10% of staffing firms currently run agentic AI across their full workflow, per Bullhorn’s 2026 GRID survey of about 2,300 recruiting professionals. (Learn more)
Why it matters: Your clients are telling investors AI is working. The Fed says most of those gains haven’t landed yet. Staffing firms that can show measurable placement outcomes already have data that most AI vendors still can’t produce.
Workers are souring on AI, and the training gap explains it
65% of knowledge workers miss how work operated before AI, and 38% would remove generative AI entirely if they could. Forty-two percent say they spend more time checking and correcting AI output than the tool saves them. Fifty-five percent say poor-quality AI output actively slows their team. (Learn more)
Only 1 in 3 workers received employer-provided AI training in the past six months, per a Conference Board survey of nearly 1,300 workers. Most of what exists focuses on basic prompting, not the skills needed to direct AI agents or catch what they get wrong. (Learn more)
Why it matters: Clients are pushing AI into workflows without preparing people for it. Workers placed into those environments carry quality and retention risks the client hasn’t accounted for. That’s worth naming in client conversations.
AI tools in hiring are landing in court
Two lawsuits filed in late July name AI notetakers used in hiring as defendants. Chamberlain v. Granola and consolidated Otter.ai litigation, both in the Northern District of California, allege candidate recordings fed into model training by default without consent. California’s Invasion of Privacy Act sets damages at $5,000 per violation or three times actual damages, with class action exposure in the millions. Speaker-identification features carry separate risk under Illinois’s biometric privacy law. (Learn more)
The EU AI Act took effect August 2, classifying all hiring, performance management, and workforce planning AI as high-risk. Penalties run to €35 million or 7% of global annual turnover. New York City’s Local Law 144 already requires annual independent bias audits and public disclosure for any AI used in hiring or promotion decisions. (Learn more)
Why it matters: Every AI tool touching a candidate interview (notetaking, screening, scoring) now carries explicit legal exposure. Agencies that govern these tools before clients ask are ahead of where the liability is landing.
The StaffingHub Brief provides weekly insights for staffing agency leaders and publishes every Friday.
