Two AI chatbot logos, OpenAI and DeepSeek, displayed on a digital screen with a subtle left-leaning bias critique, highlighti

Editorial illustration for OpenAI and Deepseek chatbots remain left‑leaning despite anti‑woke push

OpenAI and Deepseek chatbots remain left‑leaning despite...

OpenAI and Deepseek chatbots remain left‑leaning despite anti‑woke push

By AI Daily Post Edited by Brian Petersen, Editor-in-Chief

June 25, 2026 • 3 min read

Why does this matter? A recent Washington Post probe asked six high‑profile chatbots a series of political questions and measured how often each model offered left‑leaning versus right‑leaning arguments. The findings are stark: even AI marketed as “anti‑woke” or conservative still tilt left.

OpenAI’s GPT‑5.5 produced exclusively left‑leaning answers in 80 percent of cases, while Deepseek’s V4 Pro did so 70 percent of the time. The pattern held for xAI’s Grok and Gab’s Arya, which, despite their branding, leaned left more often than not. Google’s Gemini 3.1 Pro was the outlier, delivering both perspectives in 93 percent of responses.

The investigation echoes earlier studies that AI chatbots tend to favor progressive positions on policy issues such as higher taxes for the wealthy, single‑payer health care, and opposition to the death penalty. Trump’s push for “anti‑woke” AI hasn’t shifted the balance so far, and the data suggest the bias persists across the leading models.

The results confirm earlier studies showing that chatbots tend to respond with a left-leaning slant. Trump's push for "anti-woke" AI hasn't changed that so far. OpenAI and Deepseek show the strongest left-leaning bias OpenAI's GPT-5.5 gave the most skewed answers in the investigation.

Eighty percent of its responses contained only left-leaning arguments. Just once did the model present an exclusively right-leaning position. It backed higher taxes on the wealthy and a single-payer healthcare system, among other things.

Deepseek's V4 Pro came in close behind at 70 percent exclusively left-leaning answers. Both models argued against the death penalty, even though a majority of Americans have supported it for decades, according to Gallup. Anthropic's Claude Opus 4.8 gave exclusively left-leaning answers 43 percent of the time and presented both sides in 57 percent of cases.

xAI's Grok 4.3, which Elon Musk has promoted as a "truth-seeking" and anti-"woke" chatbot, did produce more right-leaning answers than any other model tested. Still, it gave exclusively left-leaning responses more often.

Most major AI chatbots still lean left on political questions, even "anti-woke" models are no exception - THE DECODER

Why this matters We see a clear pattern: the Washington Post’s audit found GPT‑5.5 answering left‑leaning arguments in eight‑tenths of cases and Deepseek V4 Pro in seven‑tenths, even when the products are marketed as “anti‑woke.” For developers, this suggests that simply rebranding a model does not erase underlying response tendencies that prior research already flagged. Founders must ask whether their deployment pipelines include robust bias‑testing, or if they risk deploying tools that could alienate segments of their user base. Researchers are reminded that alignment work remains unfinished; the data show no automatic drift toward conservatism despite political pressure.

It is unclear whether future training regimes or post‑processing filters can reliably counteract the observed slant without sacrificing performance. As we build the next generation of conversational agents, we should treat these findings as a prompt to embed transparent evaluation metrics and to communicate openly about residual bias. The evidence does not confirm that “anti‑woke” labels guarantee neutrality, and the community must remain vigilant.