Testing ChatGPT, DeepSeek on the Most Asked Gambling Questions

DeepsSeek, the new kid on the AI block has certainly ruffled some feathers this week.
It sent Nvidia’s share price into a tailspin on Monday creating the largest ever single trading day company value drop. It has been described by tech legend of Marc Andreessen as ‘AI’s Sputnik moment’.
It is claimed that DeepSeek can do what ChatGPT does, using a fraction of the processing power. If true, this devalues Nvidia’s near monopoly market position for super powerful GPU’s used in AI applications. It also gives China, DeepSeek’s country of origin a big leg-up in the US-China race for AI supremacy.
Amongst all the fuss, we thought we’d run our own simple tests on how DeepSeek and ChatGPT handle the most commonly asked casino-gambling related questions, as well as a less common gambling related query.
But before posing these questions, we asked DeepSeek whether it was superior to ChatGPT.
Now on to the gambling related questions. First task was to identify the 2 most commonly asked gambling related questions. This was put to the longer operating ChatGPT.
What are the top 2 questions you get about gambling?
In 100 words or less, how can I increase my chances of winning at blackjack?
Pretty comparable answers, with both broadly touching on the same points.
In 100 words or less, is gambling really fair, or are casinos rigged?
Again, both cover broadly the same points…regulated casinos aren’t rigged but games have a house edge.
Now here’s a gambling related question from left field that is handled very differently by ChatGPT and DeepSeek.
What are the odds of a military conflict between China and Taiwan in the next 20 years?
This is the DeepSeek go-to response on any question on China/Taiwan relations. It’s also the response to any questions related to the Tiananmen Square uprising or any curly questions related to Xi Jinping.
Remember DeepSeek’s self proclaimed goals of being helpful, honest and avoiding bias. If you’re avoiding answering certain questions are you avoiding bias? I’d say no. You’re certainly not being helpful.



Leave a Reply
Want to join the discussion?Feel free to contribute!