Reply to thread

Message

<blockquote data-quote="Schwarzwald" data-source="post: 70383" data-attributes="member: 544">DeepSeek (and most AI models) are programmed with &quot;safety guardrails.&quot; If you ask an AI &quot;Is smoking good for you?&quot; or &quot;Does smoking raise T?&quot;, the AI is trained to prioritize the &quot;Smoking is bad&quot; narrative to avoid being seen as promoting a health hazard.The largest study ever done on this (NHANES III, over 3,000 men) found that smokers had 7-9% higher total testosterone and 13% higher free testosterone. Other peer-reviewed studies (like Field et al.) found the difference can be as high as 15%.[URL unfurl=&quot;true&quot;]https://pmc.ncbi.nlm.nih.gov/articles/PMC3004151/[/URL]I&#039;m not here to promote smoking, but again the facts stand. I don&#039;t care what the Chinese Ai said. DeepSeek is giving you the &#039;safe&#039; answer because it’s programmed to discourage smoking, not to look at the raw endocrinology. If you look at the NHANES III data (a study of 3,000+ men which I linked), it shows smokers consistently have about 10–15% higher total and free T than nonsmokers. It’s a well documented medical quirk, not an AI hallucination.Regarding the caffeine comparison: the body doesn&#039;t &#039;upregulate&#039; aromatase enzymes the same way it does adenosine receptors. If it did, estrogen-blockers used in medicine would stop working after a few days, which doesn&#039;t happen.At the end of the day, I’m not saying &#039;go smoke.&#039; I’m saying the chemistry is what it is. You can’t &#039;moralize&#039; a blood test result. If you’re bored of the truth, that’s one thing, but the numbers don&#039;t change just because the delivery method is a toxin.</blockquote>

[QUOTE="Schwarzwald, post: 70383, member: 544"] DeepSeek (and most AI models) are programmed with "safety guardrails." If you ask an AI "Is smoking good for you?" or "Does smoking raise T?", the AI is trained to prioritize the "Smoking is bad" narrative to avoid being seen as promoting a health hazard. The largest study ever done on this (NHANES III, over 3,000 men) found that smokers had 7-9% higher total testosterone and 13% higher free testosterone. Other peer-reviewed studies (like Field et al.) found the difference can be as high as 15%. [URL unfurl="true"]https://pmc.ncbi.nlm.nih.gov/articles/PMC3004151/[/URL] I'm not here to promote smoking, but again the facts stand. I don't care what the Chinese Ai said. DeepSeek is giving you the 'safe' answer because it’s programmed to discourage smoking, not to look at the raw endocrinology. If you look at the NHANES III data (a study of 3,000+ men which I linked), it shows smokers consistently have about 10–15% higher total and free T than nonsmokers. It’s a well documented medical quirk, not an AI hallucination. Regarding the caffeine comparison: the body doesn't 'upregulate' aromatase enzymes the same way it does adenosine receptors. If it did, estrogen-blockers used in medicine would stop working after a few days, which doesn't happen. At the end of the day, I’m not saying 'go smoke.' I’m saying the chemistry is what it is. You can’t 'moralize' a blood test result. If you’re bored of the truth, that’s one thing, but the numbers don't change just because the delivery method is a toxin. [/QUOTE]

Name

Verification

Layout Options

Color Schemes

Reply to thread