A number published in the Dice Tech Jobs Report this month stopped a lot of people mid-scroll: AI skill requirements now appear in 75% of U.S. tech job postings. That's up from 73% in May — and up 178% compared to the same month last year. In twelve months, AI went from a differentiating requirement to a near-universal baseline.
For engineering leaders and hiring teams, this shift creates a practical problem. The demand signal is clear. The supply hasn't moved at the same pace. And the gap between "AI required in job posting" and "genuine AI capability in candidate" is wide enough to define hiring strategy for the rest of 2026.
What's driving the number
The 75% figure reflects two overlapping forces. The first is genuine: companies are integrating AI into more of their engineering work, and they need people who can operate in that environment — whether that means building models, working with AI-assisted tooling, or maintaining systems that incorporate AI components.
The second force is less precise. AI has become a catch-all term in job descriptions, often added to postings without a clear definition of what's actually required. A backend engineer who uses GitHub Copilot and a machine learning engineer building production recommendation systems are both being asked for "AI experience" — but these are fundamentally different requirements.
This conflation creates a signal problem. Companies post AI requirements. Candidates claim AI experience. Neither side is necessarily wrong, but they may not be talking about the same thing. The result is a matching problem that shows up as unfilled roles and misaligned hires.
The three tiers of AI capability that actually matter
For hiring teams trying to make sense of this, it helps to think about AI capability in three distinct tiers — and to be explicit about which one a given role actually requires.
AI-fluent engineers. These are engineers who use AI tools effectively as part of their workflow — code generation, testing, documentation, research. This is increasingly table stakes for senior engineers across all specialties, and it's appropriate to screen for it. But it's not the same as building AI systems.
AI-integrated engineers. These engineers can work with AI components in a production environment — integrating APIs, fine-tuning models, working with retrieval-augmented generation, managing the operational complexity of AI in deployed systems. This is a more specialized capability and requires direct hands-on experience.
AI-native engineers. These are the specialists: ML engineers, data scientists, and AI researchers who can design, train, and evaluate models from the ground up. The talent pool here is small, the compensation is high, and the search requires a fundamentally different approach.
Most companies posting "AI required" need tier one or two. Searching for tier three when you need tier one wastes months and loses strong candidates. Clarifying which tier the role actually requires is one of the highest-leverage things a hiring team can do right now.
Why this is making hiring harder — and what to do
The 178% year-over-year increase in AI requirements means that a large portion of the candidate pool is now being filtered through a criterion that didn't exist at scale twelve months ago. Candidates who were strong matches for similar roles in 2025 may not appear to qualify for equivalent roles in 2026, even when their underlying capability is unchanged.
This creates a few practical problems for hiring teams:
- Experienced engineers are being screened out by keyword matching. A senior backend engineer with ten years of experience who doesn't use the right AI terminology in their resume may not surface in searches designed to find "AI experience." The capability may be there; the vocabulary may not be.
- AI claims are hard to verify. Unlike years of experience with a specific framework, AI capability is difficult to assess from a resume. The gap between self-reported and actual AI competence is significant, and standard interview processes often don't close it.
- The goalposts are moving. What counts as meaningful AI experience in 2026 is different from what it was a year ago. Hiring criteria calibrated on last year's market may be filtering for the wrong things.
What the data says about what's working
The same research that produced the 75% figure also found that 70% of tech leaders say the AI factor has made them more likely to turn to staffing and consulting firms — primarily to find candidates with specialized AI skills, identify AI-related skills gaps, and manage the increased volume of AI-generated applications. And 93% of those who have worked with staffing firms on AI-related hiring challenges say the partnership has been effective.
This tracks with what AWWCOR sees in practice. The companies making the most progress on AI hiring right now are the ones who have been explicit about what they actually need, use evaluation processes designed to test real AI capability rather than surface-level familiarity, and source proactively from global talent networks rather than waiting for the right resume to appear.
The bottom line
When 75% of job postings require AI skills but the definition of "AI skills" spans a spectrum from power-user to researcher, the number itself tells you less than it seems. The companies that hire well in this environment are the ones who get specific — about what they need, how they'll evaluate it, and where they'll find it.
AI fluency is becoming a baseline expectation in tech. That's a structural shift, not a trend. The hiring processes built for a pre-AI market need to catch up — not just by adding AI to job descriptions, but by rethinking how capability is defined, sourced, and assessed.