Entry-Level Developers: The Numbers Contradict Each Other, So Which Ones Do You Believe?
Since 2025 some articles say AI is wiping out entry-level developer jobs and others say the opposite. This post lays the numbers side by side, explains why they disagree, gives six questions for reading any figure, and lists what a newcomer can do whichever side is right.
For students, newcomers and anyone looking for a first job in software. No statistics background needed. Every hard word has a short explanation.
This is a reflection: I weigh evidence and say what is unknown, it is not a research report. I read the PDFs of two papers (Stanford, Harvard) and the original pages of SignalFire, Ramp, Yale and Indeed (through FRED). I read the rest through summaries or news coverage and I say so each time. Open the original before quoting any figure.
Quick summary#
- Many articles say AI is narrowing the door into the profession, many others say nothing visible yet or the reverse. Most are right about what they measure. The trouble is that they measure different things.
- What the large studies share: no job loss across the whole economy, but employment of 22 to 25 year olds in AI-exposed jobs is weakening (Stanford, Harvard). The Stanford update (data to June 2026) gives 19%, and the authors themselves call it an early descriptive indicator, not causal proof.
- The same group from the same source still gives several numbers: 6%, 13%, 19%, 20%. They differ in baseline and method, not because someone is lying.
- Some explanations have nothing to do with AI: interest rates rising from 2022, the US tax rule on research and development costs (Section 174), and over-hiring in 2021 to 2022. Researchers tried to strip them out, but nobody has stripped them out completely.
- Before believing a figure, ask six questions: who was counted, what was counted, compared with when, where the data comes from, what else changed at the same time, how sure it is.
- Whichever side is right, a newcomer can still measure their own market, widen the doors, build real proof of skill, and re-check the numbers every quarter.
Hard words, plain meaning#
| Term | Plain meaning |
|---|---|
| Employment (headcount) | The number of people who have a job at one moment. Like counting how many people are sitting in a room. |
| Hiring | The number of people newly taken on during a period. Like counting how many people just walked through the door. |
| Job postings | The number of "help wanted" notices. Like counting signs hung outside, which does not mean anyone was hired. |
| Unemployment rate | Of the people looking for work, the share who have none. People who stop looking are not counted. |
| Baseline | The starting point used to say "fell by how much". Change the baseline and the number changes. |
| Control group | A group to compare against, showing "what would have happened without this". |
| Correlation and causation | Two things moving together does not mean one causes the other. |
| Confidence interval | The range where the true number probably lies. The smaller the sample, the wider the range. |
| Payroll data | Records of people actually paid by firms that use one payroll service, for example ADP. |
| Online résumé data | Career profiles people post themselves, for example on job networks. Wide coverage, but self-reported. |
1. Picture it: five people weighing the same crowd#
Picture it as an everyday scene. This is my own illustration, not real data.
Five people stand in front of the same crowd at a train station gate and each tries to answer: "Is this crowd getting smaller?"
- The first person counts only children. He sees fewer.
- The second counts everyone. She sees almost no change.
- The third stands at the entrance and counts newcomers. She sees a sharp drop.
- The fourth reads the notice board saying "more people needed". He sees the board emptying.
- The fifth asks each person: "Was it hard to get in?" She hears many say yes.
All five are honest and all five are right about what they measured. If you only hear "the crowd is shrinking" you cannot tell who said what. The story of entry-level jobs in software is the same, with one more twist: people also pick a different starting year and a different place to stand.
2. The side saying "the door is narrowing"#
I sort the studies by what they measure. The last column is how sure I am that I read the study correctly, not how sure I am of its conclusion.
| Source | Who is counted | What is counted | Baseline and window | Number | Data | How sure I read it right |
|---|---|---|---|---|---|---|
| Stanford, "Canaries in the Coal Mine" (Brynjolfsson, Chandar, Chen) | People aged 22 to 25 in the most AI-exposed occupations | People employed | Nov 2022 to Jun 2026 | Employment about 19% below where it would be had it kept pace with the least-exposed group (August 2025 version, data to Jul 2025: 13% after adjusting for firm-level shocks) | ADP payroll data, millions of US workers | High (read the PDF) |
| ADP Research (same team) | The youngest software developers | People employed | From the late-fall 2022 peak to Jul 2025 | About 20% below the peak | ADP payroll data | Medium (the page says "youngest software developers" without stating the age band) |
| Harvard, Hosseini and Lichtinger | "Junior" employees at firms that posted a "GenAI integrator" role | People employed, versus firms that did not post | From Q1 2023, after six quarters | Down about 9%; about 8% after eight quarters in a second test | Online résumé data on about 65 million workers at over 280,000 firms, 2015 to 2025, plus job postings | High (read the May 2026 PDF) |
| SignalFire, State of Talent 2025 | New graduates at the 15 largest tech companies by market value | People newly hired | 2023 to 2024, and versus 2019 | Down 25% from 2023 and over 50% from 2019; new graduates are only 7% of hires | The fund's own data (Beacon AI platform, over 650 million profiles) | High on the reading, low on re-checking (data is not public) |
| Indeed Hiring Lab (through FRED) | All software development roles, US | Job postings | Index Feb 2020 = 100 | From 233.82 (28 Feb 2022) to 61.10 (17 May 2025), now 80.79 (23 Oct 2026) | Postings on Indeed | High (read the original series) |
Two things to take from the table:
- The large studies do not say AI is erasing jobs across the economy. The Stanford paper itself says it finds no widespread job displacement in the economy as a whole. They describe a narrow group: young people in occupations where AI can replace part of the tasks.
- The same group and source still give several numbers. For 22 to 25 year olds in highly exposed jobs, the ADP Research page gives a fall of about 6% (late 2022 to Jul 2025, against themselves), the Stanford paper gives 13% and later 19% (against another group), and software developers alone are about 20% (against a peak). Section 4 recomputes these comparisons so you can see why they differ.
3. The side saying "not yet" or "the opposite"#
| Source | Who is counted | What is counted | Baseline and window | Number | Data | How sure I read it right |
|---|---|---|---|---|---|---|
| Yale Budget Lab (with Brookings, Oct 2025, updated 2026) | The whole US labor market | Occupational mix and length of unemployment by AI exposure | 33 months after ChatGPT launched | "No discernible disruption" economy-wide; the early-2026 update also describes stability | US public employment data (details not confirmed) | Medium (read through Fortune and news, could not open the full original page) |
| Ramp and Revelio Labs (working paper, 30 Jun 2026) | Over 21,000 US firms that spend heavily on AI (top third by spend, about 30 USD per employee per month at first) | People employed | Two years after starting to use AI | Total headcount up 10.2%; entry-level headcount up 12%; low spenders show no significant change | Ramp card spending joined to Revelio workforce records | High (read Ramp's page) |
| New York Fed, recent graduates | People aged 22 to 27 with a bachelor's degree | Unemployment rate | Q2 2026; by major, 2023 ACS data | All majors: 5.6% unemployed, 42% underemployed; Computer science 6.1%, Computer engineering 7.5% (2023, small sample, the computer engineering confidence interval is about 4% to 11% by EIG's calculation) | American Community Survey (ACS) | High for the all-graduates figures; low for by-major figures (read through an EIG analysis, did not open the original table) |
A few cautions so you do not misread:
- Ramp says "entry-level headcount up 12%", and that figure is not misreported: it is the growth in the number of entry-level people at heavy AI spenders over two years. But Ramp itself notes these firms were already larger, more engineering-heavy, more venture-backed and faster-growing beforehand. The result is a correlation among those firms, says nothing about the whole economy, and the paper is not final.
- Harvard and Ramp really do disagree about newcomers, even though both look at firms using AI. According to a note in the Harvard paper, both use Revelio résumé data. The difference is how "firm using AI" is defined: one means posting a role to integrate AI, the other means spending money on AI tools. Same data vendor, different definition, different result.
- The New York Fed by-major unemployment figures are very noisy because of small samples. The 2023 data also sits before most of the period the studies above cover. Later updates are reported by the press at about 7.0% and 7.8%, but I could not open the original table to confirm, so I do not use them as the main figure.
4. Why one market gives different numbers#
Picture it as a toy town. The numbers below are an illustration I built and computed by hand. They do not measure a real market.
The toy town has 100 positions for newcomers. In 2022: 40 positions are in occupations where AI can replace part of the work (group A), and 60 are in other occupations (group B). In 2025, group A has 32 and group B has grown to 66.
| Way of counting | Calculation | Result |
|---|---|---|
| Look at group A only, against itself | 32 / 40 - 1 | down 20% |
| Group A against the pace of group B | (32 / 40) / (66 / 60) - 1 = 0.8 / 1.1 - 1 | 27.3% lower than it would be had it kept pace |
| The whole town | (32 + 66) / 100 - 1 | down 2% |
Same data, three answers: "down 20%", "27% lower", "almost unchanged". Nobody is wrong. That is also how the Stanford paper gets its 19%: employment in the two most exposed groups of occupations fell about 11% from Nov 2022 to Jun 2026, while the three least exposed groups rose about 10%, and 0.89 / 1.10 - 1 gives about 19% below where it would be had it kept pace. If you only look at "down 11%" you get a different number from the same data.
Besides the method, four more things make figures differ:
| What differs | Example | Why the number changes |
|---|---|---|
| What is counted | People employed, hires, postings, unemployment rate | A firm can hire fewer new people yet keep its existing staff, so headcount falls more slowly than hires. People who stop looking for work are not in the unemployment rate. |
| Time window | From the 2022 peak, from before the pandemic, from the launch of ChatGPT | See the table below. |
| Who is counted | The whole country, firms using AI, new graduates | Firms using AI are larger and different from others (both Harvard and Ramp note this). |
| Data source | Payroll, online résumés, postings, surveys | Each source covers part of the market. The US CPS survey has only about 23 to 48 software developers aged 22 to 25 per month, according to the Stanford paper, so its figures jump around. |
Change the baseline, change the story: real job-posting data#
Indeed Hiring Lab's series of software development job postings in the US (index, Feb 2020 = 100, seasonally adjusted) is published on FRED. I recomputed the comparisons from three points in the series.
| Starting point | Value then | Value on 23 Oct 2026 (80.79) | Change |
|---|---|---|---|
| Peak, 28 Feb 2022 | 233.82 | 80.79 | down 65.4% |
| Before the pandemic, Feb 2020 | 100 | 80.79 | down 19.2% |
| Trough, 17 May 2025 | 61.10 | 80.79 | up 32.2% |
One series, three sentences: "postings fell by two thirds", "still about one fifth below pre-pandemic", "up 32% from the bottom". And note that in 2022 postings were about 2.3 times the pre-pandemic level. Part of the "decline" is coming back from an unusual peak, not everything is new.
5. Explanations that have nothing to do with AI#
It can be thought of as a market that went through several changes at once, with AI only one of them.
| Explanation | What happened | Evidence and confidence |
|---|---|---|
| Rising interest rates | The Fed raised rates from near zero in early 2022 to about 5.00 to 5.25% by May 2023 | Expensive borrowing makes firms hire less, especially tech firms that spend heavily on growth. The Stanford paper tried controlling for each occupation's sensitivity to interest rates and the gap remained, and it kept widening to 2026 after rates had peaked. Confidence: medium (I relied on press tracking, check the Fed's own page). |
| Section 174 tax rule | From tax year 2022 the US required companies to spread domestic research and development costs, including software development, over 5 years instead of deducting them at once. The 2025 law (section 174A) restored immediate deduction for domestic costs from tax years beginning after 31 Dec 2024 (sources differ on the start year) | Some writers argue this made firms cut staff. It is a hypothesis, no study has isolated it, and many supporting pieces are opinion. The Stanford paper has a footnote: the gap remains when tech firms and computer occupations are dropped, which speaks against this channel for their result. Confidence: low to medium. |
| Over-hiring in 2021 to 2022 | Indeed's postings series reached 2.3 times its pre-pandemic level in early 2022 | Hiring less afterwards is predictable. The peak figure is real (FRED). How much of the fall for newcomers it explains has not been separated. |
| Post-pandemic correction | Demand for digital work surged, then cooled | Same reasoning. According to news coverage of the Yale Budget Lab, many shifts in the mix of jobs began in 2021, before AI spread. |
| "Blaming AI" for cuts | Some firms say they cut staff because of AI when the real reasons are costs, tariffs or policy | Yale's Martha Gimbel calls it "AI washing", per Fortune (Feb 2026). Oxford Economics, in the same article, found 55,000 cuts (4.5% of those recorded) in the first 11 months of 2025 attributed to AI, against 245,000 attributed to market conditions. Confidence: low (read through the press). |
The Stanford paper also states its own weak points: the gap shrinks when education is controlled for, some divergent trends already appeared before generative AI, and it is more pronounced in ADP data than in national surveys. They stress that these are early descriptive indicators, not causal estimates. The literature review in the Harvard paper reports that some other studies find no systematic difference between more and less exposed occupations, for example Chandar (2025) and the Financial Times (2025) on CPS data, and a Swedish study finding a smaller fall of about 5.5%. I read these through Harvard's review, not through each original paper.
6. Six questions before believing a figure#
Picture it as six flashlights. You shine them one by one on the number, and where it stays dark, do not rush to believe it.
| # | Question | Be careful if you see this | Example in this post |
|---|---|---|---|
| 1 | Who is counted? | "Young people", "developers" in general, with no age, occupation, firm or country | Stanford covers only 22 to 25 year olds in the most exposed occupations, in the US |
| 2 | What is counted? | Unclear whether it is people employed, hires, postings or a rate | SignalFire counts hires of new graduates, not people employed |
| 3 | Compared with when, over how long? | No baseline, or a baseline right at a peak or a trough | Indeed: down 65%, down 19% or up 32% depending on the baseline |
| 4 | Where does the data come from, who publishes it? | Private data nobody can re-check, a small sample, a seller of a product | SignalFire is one fund's private data; the New York Fed by-major figures have a 4% to 11% confidence interval |
| 5 | What else changed at the same time? | Mentions only AI, not interest rates, tax, over-hiring or the pandemic | Section 5 |
| 6 | How sure: correlation or causation, which version? | "AI caused it" with no control group; the number changes by version | Stanford: 13%, then 16%, then 19% across data versions; they themselves say it is not causal |
Try it. Illustration, a headline I made up, not a real article: "Jobs for new devs fall 20%". Six questions: new devs in which country? People employed or hires? Down from which year, a peak or before the pandemic? Payroll data or postings? Were interest rates, tax and over-hiring considered? Is there a comparison group? If the article cannot answer three or more, treat the number as a rumor with a digit, not yet as data.
7. Vietnam and Japan: what I found#
- Vietnam. I found no government data specifically on new-graduate developers, and I could not open any public report from the large job platforms. The closest is a ManpowerGroup survey for Q4 2026, as reported by Tuổi Trẻ (9 Sep 2026): of 285 employers surveyed, 57% said they had increased hiring of entry-level workers compared with 2025, and 77% in the information sector. Read it with care: it is a self-reported employer survey by a staffing company, the article does not say whether the 285 are all in Vietnam, does not define "entry-level" and does not separate out developers. Confidence: low to medium.
- Japan. I found no primary data on entry-level developers in Japan tied to AI. Articles often cite a forecast from the Ministry of Economy, Trade and Industry (METI) that Japan could be short of roughly 160,000 to 790,000 IT workers by 2030 depending on the scenario. I only read it through summaries because the original page refused access, and it is a forecast of the total made before generative AI spread, so it does not answer the question about newcomers and AI. Confidence: low. Open the original before quoting.
Plainly: most figures in this post are American. Vietnam and Japan may differ, and I have no evidence to say how.
8. What a newcomer can do whichever side is right#
Picture it as not knowing whether tomorrow is rain or sun. If you carry both an umbrella and sunglasses, you do not need to guess right. The suggestions below are my own reasoning, not tested on many people.
| If the "door is narrowing" side is right | If the "nothing to worry about" side is right | What works in both cases |
|---|---|---|
| Fewer positions for newcomers, more competition | The market is only recovering after a correction | Count your own market: each week record 20 postings (role, level, requirements, whether AI is mentioned). It takes 30 minutes and gives you data where you will actually apply. |
| The title "junior developer" appears less | That title is still common | Aim at several kinds of doors: testing, technical support, data, internal IT, small firms, outsourcing companies. This is my suggestion, not a conclusion from the data. |
| Employers hesitate over people they cannot verify | Employers still need fast learners | Build real proof: a few small projects you can explain, live debugging, real contributions. See the 12-month plan for freshers. |
| Interviews get stricter | Interviews are normal | Practice interviews steadily: see the two-week plan. |
| AI does the simple work for you | AI only assists | Use AI to learn, not to do the work: write first, ask for explanations after, and still own the result. See Who is accountable when AI is wrong? |
| The numbers get worse | The numbers improve | Re-check each quarter with the six questions, and decide in advance what would make you change your plan. |
One more thing: a national average is not a verdict on one person. A single rejection usually has specific reasons, such as budget or a better-fitting candidate, not a statistic. If you want the long view, Dev career through the eras puts this in the history of the profession, and When AI writes code faster, is programming still a creative craft? covers the work that is hard to replace.
9. What is not known#
- Cause. Nobody has separated AI from interest rates, tax, over-hiring and the post-pandemic correction. The Stanford and Harvard authors each say so in their own way.
- Slower hiring or no hiring. If it is only later hiring, newcomers still get in after a few years. If a rung of the ladder has gone, that is a different story. Current data cannot tell them apart.
- Software specifically. The best figure for developers aged 22 to 25 is the ADP 20% figure and small surveys. The CPS has very few people in this group per month.
- Outside the US. I have no reliable evidence for Vietnam or Japan.
- Figures change by version. The Stanford number moved from 13% to 16% to 19% as data was added and the method changed. This post will age, so check the date.
- My own limits. I read two papers directly as PDFs and the original pages named above, and the rest through summaries. I have no raw data to re-check, so my calculations only rework published numbers.
10. A balanced view#
- This post does not tell you to worry or not to worry. The data shows one narrow area (young people in jobs AI can partly replace) weakening, and does not show an economy-wide problem.
- "Not seen yet" is not "not there". The Yale Budget Lab says it has not seen disruption, not that none will come. In the summary I read, they think AI is likely to leave a mark later.
- "Seen" is not "caused by AI". Every figure about newcomers above is a correlation.
- What can be measured is easy to turn into a story. "AI took your job" is neater than "several things changed at once and nobody has separated them". Neater is not truer.
11. Conclusion#
When two articles give opposite numbers, they are usually weighing two different things on two different scales. Ask six questions: who, what, since when, which data, what else changed, how sure. For a newcomer, doing what helps in every scenario matters more than guessing the right side: measure your own market, widen the doors, build real proof of skill, and re-check every quarter.
Start this week: open a spreadsheet, record 20 postings where you want to apply, and do not change your plan because of one headline.
References#
Open the original before quoting any figure. Where I only read a summary or press coverage, it is stated above.
- Brynjolfsson, Chandar, Chen: Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of AI (Stanford Digital Economy Lab; the version I read is dated August 2026, data to June 2026)
- ADP Research: Yes, AI is affecting employment, here's the data (the 6% and 20% figures for the youngest software developers, Jul 2025)
- Hosseini, Lichtinger: Generative AI as Seniority-Biased Technological Change (SSRN 5425555; version of 6 May 2026)
- SignalFire: State of Tech Talent Report 2025
- Ramp Economics Lab: Companies hire more after AI adoption (30 Jun 2026)
- Federal Reserve Bank of New York: The Labor Market for Recent College Graduates
- Economic Innovation Group: A viral chart on recent graduate unemployment is misleading (re-reading the 6.1% and 7.5% by-major figures, ACS 2023)
- Yale Budget Lab: Evaluating the Impact of AI on the Labor Market: Current State of Affairs and Fortune: coverage of the Feb 2026 update and "AI washing"
- FRED: Software Development Job Postings on Indeed in the United States (Indeed Hiring Lab)
- Baker Tilly: Domestic research expenditures, Section 174 under the 2025 tax act (an accounting firm's summary, not the statute)
- Tuổi Trẻ News: Vietnam among region's strongest hiring markets, yet employers struggle to find skilled workers (9 Sep 2026, ManpowerGroup survey)
- Related posts in this blog: Fresher in the AI era: a 12-month plan, Interviews in the AI era: a two-week plan, Who is accountable when AI is wrong?, Dev career through the eras
Frequently asked questions
- Is AI reducing jobs for new software developers?
- There are signals but no firm conclusion. A Stanford study using ADP payroll data found employment of 22 to 25 year olds in the most AI-exposed jobs about 19% below where it would be had it kept pace with less-exposed peers (data to June 2026). The authors call this an early descriptive indicator, not a causal estimate, and the Yale Budget Lab has not found economy-wide disruption.
- Why do studies of entry-level jobs give opposite results?
- Because they count different things. Each study picks a different group (the whole country, firms adopting AI, new graduates), a different thing to count (people employed, hires, job postings, unemployment rate), a different time window and a different data source (payroll, online résumés, surveys). The same market can produce different numbers.
- Are interest rates and the Section 174 tax rule the real cause instead of AI?
- They may contribute, but nobody has fully separated them. From 2022 the US required companies to spread research and development costs, including software, over five years instead of deducting them at once, and interest rates rose sharply from 2022. The Stanford study tried controlling for interest-rate exposure and dropping tech firms and the gap remained, which is indirect evidence, not proof.
- What should a newcomer do while the numbers conflict?
- Do things that help whichever side is right: count job postings in your own market every week, aim at several kinds of roles rather than one title, build proof that you can explain and debug your code, and re-check the numbers each quarter with the six questions in this post. These are the author's suggestions, not tested on many people.