We try to publish these monthly-trend posts grounded in actual data releases, not vibes — so instead of a generic "what's hot in tech" roundup, here's what the current numbers say, and what they mean for a business actually planning technology spend or hiring right now.
Gartner Keeps Revising Its 2026 Forecast Upward
The most telling data point this month isn't a single number — it's the pattern across three consecutive updates. Gartner forecast worldwide IT spending at $6.15 trillion (10.8% growth) in February, revised that to $6.31 trillion (13.5% growth) in April, and revised it again to $6.37 trillion (14.2% growth) in its July 27 update. Forecasters underestimating AI-driven spending three times running, in the same direction, each time, is itself a more reliable signal than any single point-in-time number — it means actual enterprise spending is consistently outpacing even optimistic projections make.
Where the Growth Is Actually Concentrated
Per Gartner's April breakdown, software spending is projected at $1.44 trillion (15.1% growth), IT services at over $1.87 trillion (the single largest category), and data center systems — the smallest category by dollar volume — is growing the fastest at 55.8%, driven by GenAI infrastructure buildout. The pattern is consistent: money is concentrating in AI-adjacent categories specifically, not spread evenly across "IT" as a whole. A business benchmarking its own tech budget against "average IT spending growth" is comparing against a blended number that undersells how much of that growth is AI infrastructure and AI-integrated software specifically.
The AI Hiring Market Is Bifurcating
On the hiring side: AI skill requirements reached 79% of US tech job postings in July, up from 75% in June — and up 144% year-over-year from July 2025. At the same time, overall tech job postings dropped 10% month-over-month in July, even though they're still up 10% year-over-year. Read together, this isn't a shrinking tech job market — it's a market where AI-adjacent skills are becoming close to a baseline requirement, and roles without that skill overlap are the ones seeing the pullback.
What This Actually Means If You're Planning a Budget or a Hire
If you're budgeting for 2026–2027 technology spend, plan for AI-integrated software and infrastructure to cost more and move faster than generic "IT spending" trend lines suggest — the growth is concentrated there specifically, and vendors in that category have real pricing power right now.
If you're hiring, AI-literate engineering talent is getting measurably more competitive and more expensive in tight markets like the US and Canada — which is a real part of the cost-saving case for an experienced offshore team that already has AI/LLM integration depth, not a separate consideration from it.
If you're evaluating whether to build an AI feature at all, rising spend on AI infrastructure doesn't automatically mean rising ROI on any specific AI project — we've written before about why measured AI productivity gains are far more uneven than the adoption numbers suggest. Spend where the data says the market is moving; don't assume every dollar spent there returns proportionally.
The Standard We Try to Hold This Kind of Post To
Every number above is dated and sourced to a named report, not a vague "industry analysts say." If a stat is repeated enough online that we can't trace it to a primary source, we leave it out rather than pass along something we can't stand behind — a "market trends" post that can't be checked against the actual reports isn't worth publishing.
This post covers the broader IT spending and hiring picture. For what's specifically happening in AI this month — big tech's fight over open-weight AI regulation, the industry's AI safety grades, OpenAI's IPO timeline, and a significant agentic AI security incident — see our companion posts below.
If you want to talk through what any of this means for your own technology roadmap or hiring plan, reach out at info@digit.com.pk — we'll ground the conversation in your actual numbers, not a trend report.