ChatGPT Is Flooding College Transcripts With A's — And Employers Are Noticing

A new study from the University of California, Berkeley has put hard numbers on what many professors already suspected: AI tools like ChatGPT are inflating college grades at a measurable scale. Analyzing more than half a million grades from 2018 to 2025 at a large public university in Texas, senior researcher Igor Chirikov found that professors in AI-exposed courses — particularly writing- and coding-heavy classes in humanities and engineering — handed out roughly 30% more A's after ChatGPT's late 2022 launch. Courses with more take-home assignments showed the sharpest spike, a telling sign that unsupervised work is where AI does the heavy lifting.

Chirikov is careful to note the grades don't reflect smarter students — they reflect smarter tools. Through 2022, AI-exposed and non-AI-exposed courses tracked nearly identically in grade distribution. The divergence afterward points squarely at generative AI as the driver. His concern is that the 'productive struggle' essential to genuine learning is being bypassed, leaving graduates who can produce polished outputs without having developed the underlying cognitive skills to do so independently.

The timing is awkward for employers. After years of de-emphasizing GPA in hiring, companies are swinging back the other way as the entry-level job market cools and applicant volumes surge. The share of employers using GPA as a hiring filter climbed from 37% in 2023 to 42% today, per the National Association of Colleges and Employers. Barclays, Morgan Stanley, and others have reinstated hard GPA minimums for some internship roles. On the career platform Handshake, nearly a quarter of GPA-requiring job postings now demand a 3.5 or higher — up sharply from just 9% in 2020.

Elite universities are sounding the alarm. A February report from Harvard College acknowledged that employers have a legitimate interest in comparing students' academic performance but concluded that current grading policies make that impossible — faculty are voting this week on a proposal to cap the number of A's awarded. Yale was even blunter in an April report, declaring that grades at Yale, as at many peer institutions, no longer communicate what students have learned. The broader irony is sharp: employers are leaning harder on GPA precisely as AI renders it less meaningful.

The corporate world is caught in its own contradiction. Research from Veris Insights shows companies simultaneously resent AI-assisted job applications and expect graduates to be proficient AI users. One Wharton instructor responded pragmatically — shrinking the weight of homework in final grades after realizing AI could score a perfect 100% on her assignments, and shifting emphasis to in-class exams where AI access is blocked. The episode captures a wider reckoning in higher education: if AI can ace the work, what exactly is a grade measuring anymore?