Adzuna’s July figures put UK graduate vacancy ads at 8,383, down 45.6% year on year and below any month in its series since April 2016. The collapse is sharper than other measures of entry-level demand, and the available evidence does not apportion it between softer hiring, higher labour costs, skills mismatches or AI.
Adzuna, the UK job-search platform that builds an index from online job ads, says its July data put graduate vacancies at 8,383—down 45.6% from July 2025 and lower than any month since the company began the series in April 2016. The figures, reported from Adzuna’s release and independently described, describe a much thinner advertised route into work for university leavers; the series exceeded 55,000 graduate vacancies in mid-2017.
The number is consequential, but it is not a count of graduate employment, applications or people unable to find work. Nor do the available data identify the share of the decline caused by AI. The sharper question is whether employers are permanently changing entry-level work, or whether a fragile hiring market has produced an unusually deep dip in one measure of advertised graduate roles.
Adzuna’s July figures show graduate ads falling far faster than its broader measures. The like-for-like comparison is useful because all three figures come from the same index and use the same July-to-July window; it does not show why the categories moved differently.
| Adzuna UK measure, July | Level | Change from a year earlier |
|---|---|---|
| Graduate vacancies | 8,383 | -45.6% |
| Non-graduate entry-level vacancies | 192,864 | -8.1% |
| All vacancies | 791,490 | -9.6% |
The reported Adzuna figures also put competition at 2.14 jobseekers per vacancy, up from 1.93 a year earlier. Andrew Hunter, Adzuna’s co-founder, called July “a step backwards, not a blip,” while saying the firm had seen gradual improvement in earlier months.
Adzuna’s own May report helps define both the series and its limits. The company says it collects UK vacancies advertised online from more than 1,000 sources, then normalises, de-duplicates, mix-adjusts and removes outliers in real time. That makes the index a measure of online advertising, not a census of all vacancies, hiring decisions or graduate jobseekers. In May, it counted 8,398 graduate vacancies, down 42.1% annually—evidence that the July result followed an already weak spring rather than appearing from nowhere.
An Indeed series cited in a separate report found postings with “graduate” in the title down about 7% year on year as of July 10, the lowest point for that point in the year since 2020. It is not a rebuttal of Adzuna’s result: the label, collection method and date differ. It is, however, a reason not to turn the 45.6% figure into a claim about every form of entry-level hiring.

Government-published LinkedIn chart shows selected entry-level occupations with annual hiring gains and declines in April 2026. Source: Department for Science, Innovation and Technology.
The sources offer several plausible pressures: a weak economy, employers’ higher labour costs, uncertainty, a skills mismatch and the use of automation for routine junior work. They do not allocate the graduate-vacancy fall among them. Reports that companies are using AI for administrative tasks establish a possible mechanism, not the size of its effect on this series.
The Department for Science, Innovation and Technology, a government department, published a brief prepared by LinkedIn and its AI & the Future of Work Unit. It found that UK hiring was down 14% year on year in April; 30 of 38 tracked entry-level occupations were declining. Accountants (-29%), graphic designers (-28%) and software engineers (-27%) were among the fastest-falling information-processing roles, while sales and customer-facing roles were growing in the brief.
That study says entry-level hiring was falling broadly in line with the wider market, unlike the especially steep Adzuna graduate series. It also says employers in shrinking occupations favour experienced candidates with operational skills that entrants often lack. The authors note that the occupations declining most sharply are those where AI has become visibly capable—but explicitly say further research is needed before drawing conclusions. The datasets report concurrent declines without establishing a single cause.

Adzuna’s company-reported chart shows annual change in hiring activity across occupation groups in May 2026. Source: Adzuna UK Job Market Report.
Adzuna said teaching, travel, trade and construction were still adding graduate roles, with manufacturing also improving. Healthcare, nursing, hospitality and logistics were among the sectors with fewer vacancies, according to the sector account. Those gains do not offset the overall graduate decline, but they complicate a story of an economy-wide disappearance of first jobs.
The exposure is also wider than graduates without being the same population. The Office for National Statistics estimated that 1.012 million UK people aged 16 to 24 were not in education, employment or training in January to March, equal to 13.5% of that age group and 89,000 more than a year earlier. The ONS release is a UK, seasonally adjusted Labour Force Survey estimate, not a graduate statistic; it includes people who are unemployed and economically inactive. The figures are official statistics in development, and the ONS cautions against reading short-term movement in smaller groups too strongly because sampling variability is higher.
That distinction matters. A fall in online graduate ads may increase the difficulty of entering work, but it cannot by itself explain the NEET total or prove that more graduates are out of work.
The next releases need to show whether the Adzuna series recovers with the broader vacancy market, whether its growing sectors continue to add graduate openings, and whether the gap with other graduate-posting measures persists. More importantly, evidence on actual hires, job redesign and the skills required—not only vacancy titles—would be needed to separate a cyclical hiring retreat from a durable AI-driven change in the first rung of professional work.
Get concise AI news and useful context from the Magica team.
Read the newsletterCompanies are expanding AI monitoring and patching as cyber threats strain security teams. Binance’s preliminary breach findings and attacks on aging refrigeration systems show why faster tools also need tighter access controls and investment in basic infrastructure.
Five technology groups have accumulated an estimated $3.2 trillion in future purchase and lease commitments. Filings show the funding pressure, but broad spending categories, long payment schedules and a methodology change limit what the total says about AI debt and growth.
PwC finds growth in a subset of AI-exposed junior job postings demanding senior skills, alongside pressure on entry-level opportunities. Workplace research suggests AI can support learning, but leaves employers to demonstrate how recruits acquire independent expertise.
CoreWeave bonds fell after the election of candidates favoring data center restrictions. Texas governor-elect Gina Hinojosa wants to extend an existing approvals pause until lawmakers pass protections, adding delivery uncertainty for an industry whose expansion depends on power and financing.
Walmart plans new permissions and a gateway for outside shopping agents, while Amazon blocks unauthorized tools and sells ads through its own assistant. The stakes are product discovery, customer data and whether delegated purchases add sales or displace advertising opportunities.
Apple reportedly cut some engineers’ daily AI allowances from $300 to $200 in October and is weighing deeper restrictions. The limits tighten control over model use, but unused allowances make the savings uncertain.
Factory is acquiring coding-agent collaboration platform HumanLayer after raising $200 million in September. It plans to integrate shared tasks and planning tools, while HumanLayer promises open source, refunds and at least 30 days of continued service.
A minister identifies Google-commissioned research behind the action plan’s £400 billion AI claim. The government’s central estimate is £120 billion, but different methods and adoption assumptions prevent a straightforward forecast-cut comparison.
Research has separated language models’ detection of altered internal activity from their ability to identify its content. Mechanistic studies and skeptical controls leave the interpretation disputed—and one intervention that improved detection made other measures of self-knowledge worse.
The IEA maps 134 policies governing data-center integration, arguing that credible project queues, coordinated tariffs and flexible connections can protect consumers while accommodating AI growth. Many measures remain too new to judge.
Three watchdog groups seek to halt Project Meridian, alleging that its contractor-heavy membership and closed meetings violate federal advisory committee law. MITRE’s invitation for outside ideas complicates the dispute over public participation.
Meta is seeking up to €9 billion in its first euro bond sale, drawing record orders while offering wider spreads than Alphabet. Its advertising business and large liquid reserves cushion an AI investment push that has overtaken quarterly operating cash flow.
Pangram is expanding its text-detection business into images and video. Browser checks and developer access broaden its reach, while company benchmark scores leave real-world error rates and costs unresolved.
About two dozen restrictive data center measures appeared to pass in initial returns. Texas governor-elect Gina Hinojosa’s promised construction pause adds a state-level constraint, though the capacity ultimately delayed remains uncertain.
Riot Platforms says its 15-year Corsicana lease with Anthropic will generate about $28 billion in revenue. The tenant owns and operates the compute, while Riot develops power infrastructure and funds module reimbursements through debt or credits against future rent.
Sonos plans limited U.S. early access for its conversational assistant, with a reported November 6 start. Users can return to older voice controls—a choice that also separates local audio processing from the new assistant’s cloud processing.
Tesla, SpaceX and AI infrastructure shares fell after Democratic election wins, while broader stocks rallied. Texas governor-elect Gina Hinojosa’s proposed data center pause puts a major construction pipeline in focus, but Republicans retain control of the Texas Legislature and federal agencies.
Capital One Ventures leads Sentra’s $150 million Series C and gains a board seat as the company expands AI-agent data controls. Sentra reports roughly 100 major enterprise customers and revenue in the tens of millions, but exact revenue, valuation and comparative costs remain undisclosed.
Google agreed to support Anthropic’s lease obligations at TeraWulf’s Kentucky campus, strengthening its bid for project financing. The support starts after delivery milestones, while debt terms and a guaranteed construction price remain unsettled.
OpenAI plans to acquire Pierre Computer and integrate its Code Storage service into Codex. The purchase would extend an existing supplier relationship into code infrastructure, but its terms, customer economics and relationship to OpenAI’s own Pando system remain unresolved.