Tagged articles

model degradation

4 articles · Page 1 of 1

Sep 15, 2026 · Artificial Intelligence

Codex 'Dumbing Down' Claim Tested: Relay Accounts vs. Temporary Downgrades

The author empirically tests viral claims that OpenAI secretly degrades Codex for relay accounts, analyzing two benchmark tasks — a candy probability puzzle and an SVG pelican animation — finding local Codex passes both, while clarifying OpenAI's documented temporary downgrades target suspicious login activity, not relay usage specifically.

CodexGPT-6 AstraLLM evaluation

0 likes · 8 min read

Codex 'Dumbing Down' Claim Tested: Relay Accounts vs. Temporary Downgrades

Java Tech Enthusiast

Apr 18, 2026 · Artificial Intelligence

Why Claude Code Is Failing Complex Engineering Tasks: AMD’s Deep Dive Reveals Four Critical Flaws

An AMD AI director’s GitHub issue sparked a data‑driven investigation that uncovered four major shortcomings, a 67% drop in thinking depth, a surge in API usage costs, and concrete recommendations to restore trust in Claude Code’s ability to handle complex engineering workloads.

AI coding assistantAMD AI directorClaude Code

0 likes · 12 min read

Why Claude Code Is Failing Complex Engineering Tasks: AMD’s Deep Dive Reveals Four Critical Flaws

Data Party THU

Oct 28, 2025 · Artificial Intelligence

Can Low‑Quality Data Cause Irreversible ‘Brain Rot’ in Large Language Models?

Researchers from Texas A&M and UT Austin demonstrate that prolonged pre‑training on low‑quality, short‑form web content causes large language models to suffer irreversible cognitive decline—manifested as attention loss, broken reasoning chains, and personality distortion—highlighting data quality as a critical training‑time safety issue.

Cognitive SafetyLLMRepresentational Drift

0 likes · 7 min read

Can Low‑Quality Data Cause Irreversible ‘Brain Rot’ in Large Language Models?

Programmer DD

Jul 21, 2023 · Artificial Intelligence

Why Did GPT-4’s Performance Plummet Between March and June 2023?

A Stanford‑Berkeley study reveals that between March and June 2023 GPT‑4’s accuracy on prime‑checking fell from 97.6% to 2.4%, code generation quality dropped sharply, and sensitivity handling changed, underscoring the rapid, unpredictable shifts in large language model performance over short periods.

AI safetyGPT-4LLM evaluation

0 likes · 6 min read

Why Did GPT-4’s Performance Plummet Between March and June 2023?