You've probably heard two completely different stories about AI at work this year. One says it's already paying for itself. The other says companies are quietly burning money on pilots that never go anywhere.

New data says both are true, just for different teams. We'll break down which functions are actually seeing a return on AI agents, why the gap is so wide, and what it means for how you talk about AI in your own job search.

Plus a manufacturing hiring pocket hiding inside a weak jobs report, a sandwich chain worth $7.9 billion, and a fossil ape rewriting where apes came from.

Let's get into it.

The June jobs report (https://www.bls.gov/news.release/empsit.nr0.htm) added just 57,000 nonfarm payrolls, a sharp slowdown from earlier in the year, with unemployment holding at 4.2%. Healthcare and social assistance carried most of the growth again, while professional services, information, and finance kept losing ground.

Retail sales rose a modest 0.2% in June, the smallest increase in five months, according to Census Bureau data (https://www.cnn.com/2026/07/16/economy/us-retail-sales-june). Core sales, which strip out volatile categories, actually fell 0.2%, the first decline in over a year, even as headline spending held up.

Monday.com became the latest company to cite AI (https://techcrunch.com/2026/07/25/the-running-list-major-tech-layoffs-in-2026-where-employers-cited-ai/) in a round of job cuts this week. It joins a running list that now covers 322 layoff events and roughly 206,000 workers in 2026, with 54% of those events citing AI or automation as a factor.

Jersey Mike's launched its IPO this week, pricing shares (https://www.axios.com/2026/07/20/jersey-mikes-blackstone-ipo-terms) between $21 and $25 and targeting a valuation as high as $7.9 billion. Reformation, the apparel brand, priced its own offering the same week at roughly $225 million, both set to start trading July 30.

Chip industry job postings are up 47% year over year in the first half of 2026, according to Seoul Economic Daily (https://en.sedaily.com/finance/2026/07/21/chip-hiring-booms-first-half-job-postings-surge-47-percent), concentrated in entry-level roles as fabs race to staff up capacity.

Funding

Contracts

Your customers are searching in places your strategy doesn’t reach.

So before your business is buried and left behind, you need to understand the new rules of SEO.

BELAY's SEO in the Age of AI report explains how search is changing, what AI means for your visibility, and the practical steps small businesses like yours can take to stay visible.

BELAY’s U.S.-based Marketing Assistants turn strategy into execution, helping your business stay visible, credible, and competitive in every search.

Download a copy for free

Companies have spent two years arguing about whether AI is actually worth what they're paying for it. New benchmark data finally puts real numbers behind both sides of that argument.

Bain's 2026 Agentic AI Benchmark found that AI agents deployed in customer service pay back their cost (https://helalabs.com/blog/agentic-roi/) in a median of 4.1 months, with 63% of customer service programs reaching payback within a year. Marketing operations takes closer to 6.7 months. Engineering is the slowest, at 9.3 months, and it's also where deployments most often stall out entirely.

Across all functions, only 41% of agent rollouts cross into positive ROI within their first year, and 19% never get there. The ones that do succeed aren't marginal wins either: Bain measured an average 171% return globally, and 192% in the US, once an agent actually works.

Gartner has been warning (https://www.gartner.com/en/newsroom/press-releases/2025-06-25-gartner-predicts-over-40-percent-of-agentic-ai-projects-will-be-canceled-by-end-of-2027) about this same split for over a year, predicting more than 40% of agentic AI projects will be canceled before the end of 2027 over cost overruns and unclear value. The pattern lines up: tasks with a clear, repeatable structure show fast returns, tasks that require judgment across a messy, shifting codebase don't.

Why it matters: If you're job hunting and a company asks how you'd use AI in the role, the strongest answer names a specific, bounded task, not a vague "I use AI a lot." Structured, repeatable work is exactly where these tools already prove their value, and being able to point to that distinction in an interview signals you understand the technology better than someone reciting features.

June's headline number, 57,000 jobs added, was one of the softest prints of the year, and the unemployment rate held at 4.2%. But zoom into the sector breakdown and the picture splits.

Manufacturing added jobs (https://resumehog.com/blog/posts/july-2026-jobs-report-breakdown-how-to-navigate-a-slow-hiring-market.html) at a pace roughly 120% higher than a year ago, even if the raw monthly count was small. Construction extended a four-month hiring streak. Chip manufacturers, meanwhile, are running entry-level job postings up 47% year over year as fabs race to staff new capacity tied to the AI buildout.

None of these sectors show up in headlines about "the jobs report." They're not glamorous, and they're not where most job search advice points people. But they're hiring while professional services, information, and finance keep shedding roles.

Why it matters: If your search has stalled in a sector that's contracting, these adjacent sectors are actively adding headcount right now, and manufacturing and construction operations roles increasingly need the same project management and process skills that consultants and strategists already have.

→ Land your next role faster with Tailr, our Claude-powered job search and networking tool: Start searching

→ If you're targeting $150K+ roles, have our team run your entire job search for you: Apply here

→ Build your own personal AI workforce handling your work, business, and personal life: Access to ready-made agents

Figure out if a target company is actually seeing AI ROI, or just talking about AI. Copy this into Claude or ChatGPT:

"Here are the most recent earnings call transcript excerpts, press releases, or job postings from [company name] mentioning AI. Based only on this text, tell me: (1) do they cite specific, measurable outcomes from AI (cost savings, headcount changes, revenue impact) or only aspirational language, (2) which business function seems furthest along, and (3) what follow-up question I should ask in an interview to test whether their AI initiative is actually delivering results. Be direct if the evidence is thin."

Last issue's result: the Anthropic copyright ruling is still too fresh to call on the "another lab settles within 6 months" prediction. We'll revisit it in the fall.