Nov 6, 2025
Can you trust ChatGPT?
Karen Hao’s Empire of AI and recent claims of Open AI’s co-founder’s statements on Altman’s “consistent pattern of lying”, reminds me of Mark Zuckerberg’s famous “dumb f****” statement; time and time again these CEOs tell on themselves and show us, while they may feign human interest and rights, that chasing profits at any cost, including selling our data wholesale, is fair game.
OpenAI, through ChatGPT, has been accused of encouraging suicide, harboring dependency with its users as well as censoring information or being used to sway public perception. In the subtlest of ways, it can just omit information, or repeat back information to you using specific framing, especially in your tone or way of speaking, acting as a mirror, which inherently builds trust.
And I am no luddite – I get it, sometimes technology goes through growing pains and makes mistakes; in fact, I understand the major complaint of GPT-5 users is that it is way duller and less personable than GPT-4o. This was done in part to protect users and prevent the misuse of ChatGPT.
Also, I m not singling ChatGPT out – ask DeepSeek any question about Taiwan and the Chinese government and you can expect a similar response. This includes TikTok, especially since Oracle’s Larry Ellison has decided to take it over, possibly at the behest Netanyahu and AIPAC identifying that the Chinese-based platform as being a critical ‘weapon’ in the “information war”.
While most of us use social media to stalk our crush’s latest updates, watch cat videos, and share videos of our children’s antics, more recently we’ve needed it to give us raw news coverage on the atrocities occurring in Gaza, Sudan and elsewhere in the world.
Now, these platforms increasingly are becoming used to manipulate narratives, and different nations are using different platforms to push their individual propaganda. X (formerly twitter) has increasingly been used to push Islamophobia, as concluded by this study from the Center for the Study of Organized Hate. Have we already forgotten about Cambridge Analytica attempting to rig US elections? 7 years later there is a $8 billion dollar lawsuit for the “data breach”.
Instagram, through its parent company Meta, has been accused of shadow banning Pro-Palestine content, including suppressing certain hashtags, shadow banning comments, removing non-graphic picture of war, among other things.
All of this clearly demonstrates a pattern – whether or not you are Pro-Palestine or Pro-Zionist, this doesn’t change the fact that those who are the stewards of the largest technological monopolies are weaponizing their platforms and influence to push agendas. How deep does this go?
You may ask “okay, states are using propaganda, did you really write 500 words to tell me water is wet? What am I supposed to do? Stop using Instagram?”
Yes, I believe you should; but how? We can’t simply just put our phones down. Whether it’s Reddit, which was a testing ground that proved AI comments are way more persuasive than human comments, or where users complain of mass censorship and bots influencing opinion, or Instagram, TikTok, X, and anything else, when we allow ourselves to stay in the realm of oligopolies, we are exposed.
The real question isn’t whether or not we should trust these systems; it’s “do we have the courage to build something better before the window closes?”. We need alternatives.
We need to develop social media platforms, tech alternatives to ChatGPT and even operating systems that allow us to keep our data, our privacy, and prevent us from being exposed and vulnerable. We need the tech folks that have the brains, the entrepreneurs with the courage, and the business folks who can take ideas to market.
Imagine, a social media platform that doesn’t farm your data or sell it to the highest bidder; a ChatGPT alternative that doesn’t build a psychometric profile on you and has predefined prompts to influence your opinion. An operating system that doesn’t have a back end for law enforcement to access your data whenever they desire. And this isn’t impossible to do.
Now, more than ever, its very easy to spin up new technologies, ironically perhaps, because of LLMs. LLMs are allowing the average person to build anything they want; this is even leading some countries like India to have whole firms dedicated to creating localized alternatives to software giants from the West, offering them with one-time fees, effectively undercutting their whole SaaS-based businesses.
This new world means that while yes, the code made by LLMs needs to be checked, the apps made secure, and business folks need to be around the table to push the technologies and operationalize them, we can effectively compete with major software vendors and form our own inexpensive alternatives.
Our window to act is closing.
The UK is rolling out Digital IDs, which have been faced with severe pushback and controversy, as many are thinking to the example of China’s Social Credit Score; where, theoretically, a person that exists outside of the social “norms” can be prevented from travel or making purchases.
With the fact that many governments, including the EU, are exploring digital currencies, more and more the writing on the wall is that our lives are going to be controlled increasingly by AI. The more we use these systems, the more you are funding them; the more you ask ChatGPT, the better it will control you.
AI that can read all of your data, your location, your personality, your interests, your relationships, your salary and bank balances, and so much more, can make a determination on who you are. Palantir has already been using data to provide the Trump Admin with data on Americans, and OpenAI is in strategic partnership with Palantir, having won a $200mm contract with the Department of Defense (War).
So, I ask you again, can we trust ChatGPT? Or is it time we start building alternatives? We are not just consumers; we are the architects of the digital future. All we have to do is choose to take ownership and be brave enough to build the world we desire.
