The Core

LLMs prefer their own writing on resumes to that of humans or other LLMs.

"The bias against human-written resumes is particularly substantial, with self-preference bias ranging from 67% to 82% across major commercial and open-source models."

https://arxiv.org/abs/2509.00462


Summary here
https://x.com/heynavtoor/status/2048088874686300431

QuoteSame person. Same resume. Same skills.

One version written by a human. One rewritten by GPT-4o.

GPT-4o picked its own version 97.6% of the time.

This matters a lot since more and more firms do an initial filter with AI. So if you know which model their HR department uses, you have a massive leg up.

Thanks for sharing.  Super interesting.

Also extending beyond the resume example.  Google is naturally against AI-scaled content for obvious reasons, but wonder to what extent Gemini assisted content generation may perform better in Google products than ChatGPT/Claude/etc.

I passed that on to a godson of mine looking for work. Thx.

works until:

QuoteEY uses its own proprietary large language model (LLM) called EY.ai EYQ to power its AI-driven recruitment and talent management processes.Key details on EY's AI recruitment tools:

EYQ: This is a secure, internal generative AI tool developed by EY that supports conversational AI assistants, talent management, and recruitment.
EY.ai Workforce: This platform combines IBM watsonx Orchestrate with EY's HR domain knowledge to automate and enhance recruitment and HR processes.
Recruitment Features: The AI systems are used for CV screening to analyze content, match keywords, and rank candidates based on qualifications, skills, and experience.
AI Training: EY also uses Azure OpenAI capabilities to power, and its teams have access to GPT-4 through partnerships with Microsoft
.The goal of these AI tools is to create a more personalized, efficient, and equitable hiring process, while ensuring data privacy for applicants.

equitable....