Do LinkedIn's AI hiring tools actually produce better hires?

The release's most concrete evidence for its AI hiring tools is about speed and volume, not the quality of hires. For Hiring Assistant, "early adopters are saving 4+ hours per role, reviewing 62% fewer profiles, and seeing a 69% improvement in InMail acceptance rates." For Hiring Pro, small businesses that report time savings see "an average of 6+ hours saved weekly," and "nearly 60% of hirers find a candidate to interview within the first week." Each figure comes from LinkedIn, describes users who chose to adopt the tool, and is given without a comparison group or baseline.

The reasoning the release gives follows the same scarcity argument as its adoption figures. Recruiters face "pressure to fill roles faster (42%)" and a need to "uncover 'hidden gem' candidates (39%)." Saving time and reviewing fewer profiles answer the first pressure directly. The second is the only one that touches quality, and the excerpt's evidence for it is again self-report: 59% of recruiters say AI "is already helping them discover candidates with skills they wouldn't have found before." That claim is that the search surfaces different people. It does not say those people become better hires, stay longer, or perform better once hired.

The nearest test of what these agents can do is the workplace benchmark (Why do AI agents fail at workplace social interaction?), which measures completion of simulated tasks and flags social interaction as among the hardest. The release reports no completion or error rates for Hiring Pro's outreach or its AI-powered interviews, so the benchmark's failure pattern cannot be checked against these products. The sibling note on adoption (Are recruiters and job seekers really adopting AI in hiring?) takes the usage figures as given; this question asks what the users get from them. The WORKBank note (What collaboration level do workers actually want with AI?) measures what workers want AI to do in their jobs. The excerpt measures no candidate preference about being screened by AI, so whether candidates want that is also open.

What the excerpt does not establish is whether AI-screened or AI-discovered candidates are better hires, whether they stay, or whether the "more valuable conversations" that 70% of recruiters expect actually happen. The release publishes no hiring-outcome data at all. Answering the question would take outcome data for hires made with and without these tools, which this release does not provide. Until that exists, the efficiency figures support a narrower claim: early adopters report saving time and reviewing fewer profiles, and nothing stronger.