Is AI evil? The Claude vs. Chatgpt debate.
- rmclements10
- Mar 12
- 7 min read
Updated: Mar 16
Let's talk about the entire concept of ethical and human-focused AI
A lot of people are cancelling ChatGPT and switching to Claude right now.
Thinking it is the more "ethical" choice.
And I know that no act of resistance is too small.
But,
also,
we need to acknowledge that Claude is not ethically superior to Chatgpt.
What's true about OpenAI/ChatGPT and Trump:
OpenAI co-founder and president Greg Brockman gave Trump's super PAC MAGA Inc. $25 million - the largest donation of the six-month fundraising cycle - widely seen as an attempt to curry favor with the Republican administration. Yahoo Finance
Beyond Brockman, OpenAI executives and top investors have collectively donated over $100 million to "Leading the Future," a super PAC that opposes state-level AI regulation in favor of a national framework. The group has already run ads against legislators who support AI safety guardrails. CNN
The Trump administration has also struck a deal with OpenAI, making ChatGPT Enterprise available to all federal agencies for $1 per agency - framed as part of Trump's goal to "win the global AI race." Fox News
What's true about Anthropic:
Anthropic announced it is giving $20 million to a political group campaigning for more regulation of AI - the opposite direction from OpenAI's political spending. CNN
Anthropic also refused a Pentagon request for unrestricted access to its AI for mass surveillance or weapons that kill without human oversight. Quitgpt
Important nuances:
These are donations and decisions by individuals and executives, not the AI products themselves. ChatGPT the product isn't "funding" Trump - specific people at OpenAI are.
OpenAI the company says it is not itself contributing to super PACs, even while its executives clearly are. CNN
Anthropic is not purely neutral either - it did sign a separate government deal and operates in the same commercial environment.
Both companies ultimately need government relationships to operate at scale.
OpenAI's leadership has made substantial, politically aligned financial decisions that Anthropic's leadership has not. Whether that reflects the companies' values or just individual executives' choices is a legitimate debate, as ultimately companies are ultimately people.
What is also important to remember about both companies.
The current design for data centers being utilized for AI in the US uses an inordinate amount of water. Water is a limited global resource. And all companies should be investing in processes to cool their data centers that do not use as much of this precious resource and should also be held accountable for investing in alternative solutions to the global water crisis.
In an earlier post I proposed the concept that we embrace AI - with guardrails and with humans at the center.
Technology will always evolve.
And AI isn't the problem in itself.
And a specific AI platform or tool isn't the problem.
Anything can be corrupted and used for evil - a calculator can be turned into a bomb, a surgical scalpel into a weapon.
Humans have to be the one to acknowledge the potential for risk and work proactively to protect against it.
Building up an "ai is evil" overreaching arguments is ridiculous and misses the point entirely. AI isn't an autonomous evil being like James Spader's Jarvis in the Avengers movie.
AI is a tool - that we are creating.
We have to create it.
You have to create it.
Are you are worried about how AI can be used to reinforce racism and sexism? You should be.
Artificial intelligence is making decisions that shape people's lives every day - who gets a job interview, who qualifies for a loan, whether a doctor's recommendation gets flagged for review.
These are NOT NEW.
I was using AI to review job applications at Frontier Airlines in 2017.
Sorry for all the LinkedIn peeps who think it's new to 2026.
AI doesn't arrive at neutral decisions. It learns from data, and that data reflects the world we've built - a world already layered with racism, sexism, and structural inequality. When we feed biased history into these systems, we don't just preserve those biases. We scale them, automate them, and make them harder to see and challenge.
MIT's Gender Shades project revealed dramatic disparities in commercial facial recognition systems - some showing error rates as high as 37% for darker-skinned women, compared to much lower rates for lighter-skinned men. Kodexo Labs
A 2025 study found that AI resume-screening tools gave older male candidates higher ratings than female candidates and younger candidates, even when all resumes were generated from identical underlying data. Crescendo
A Cedars-Sinai-led study found that leading AI models - including several of the most widely used - generated less effective treatment recommendations when a patient's race was identified as African American. Crescendo
This isn't a bug. None of this is suprising and people have been calling out these systemic flaws for decades. They are features of systems built without equity in mind.
AI is compounding the pre-existing concerns because of it's scale.
We we currently actively shaping the future of our global society right now. And AI is being used to rapidly scale some of these systems across industries and countires. Now we have those inequitable systems being applied at mass. That is the biggest problem.
Why This Happens
AI systems learn to perform tasks from the data they're trained on. When those datasets carry conscious or unconscious bias - like showing men as scientists and women as nurses instead of doctors - the AI learns that pattern and replicates it in decisions about hiring, lending, and legal judgment. ONU Mujeres The danger is that unlike human bias, which can be inconsistent and challenged in the moment, algorithmic bias creates reproducible patterns of unfairness that operate at scale, affecting thousands or millions of decisions simultaneously. Kodexo Labs
When AI systems are developed without adequately accounting for existing racism, sexism, and other inequities, built-in algorithmic bias can result in invisible but very real discrimination. As these systems are deployed, they exacerbate existing disparities and create new roadblocks for already marginalized groups. American Civil Liberties Union
How You Can Fight Back
Use AI.
And hold it accountable.
Don't be afraid of it.
Everyone has free courses right now on how to utilize AI in every aspect of your life. Use it and call it out when it gives you data or recommendations pulled from biased sources. Ask for sources. Check to see how varied they are. Share your experiences online and connect with other people who are noticing trends that appear to be biased.
Build your own AI agent that is actively sourcing from varied sources to provide equitable and unbiased information.
Learn to recognize it.
AI bias is often invisible by design.
Start by asking: what decisions in your life or workplace are being made by algorithms? Hiring platforms, credit systems, healthcare tools, and criminal justice risk assessments are all places where bias quietly operates. Naming it is the first step.
Start by asking: what decisions in your life or workplace are being made by algorithms?
Demand transparency and audits.
New York City already has a law requiring annual bias audits for automated employment decision tools and public reporting of the results. California finalized similar regulations in late 2025, and Colorado's AI Act, effective June 2026, will require developers and users of AI hiring tools to use reasonable care to prevent algorithmic discrimination. Sanford Heisler Sharp McKnight
Push your local representatives for the same where you live. You should already be contacting them about so many other things - just add this to the list.
Diversify who builds AI.
One of the most durable fixes is structural: the people designing these systems need to reflect the full range of people affected by them. Support scholarships, apprenticeships, and initiatives that bring underrepresented communities into AI development, not as an afterthought but at the foundation.
Use your wallet and your voice.
Organizations deploying biased systems may face legal repercussions, ethical dilemmas, and financial penalties. EY Consumer pressure works. Ask the companies whose products you use what their bias auditing practices look like. The Claude vs. Chatgpt showdown the last week is the perfect example of public and financial influence.
Organizations Leading the Way
Algorithmic Justice League (AJL) — Founded by MIT researcher Joy Buolamwini, whose Gender Shades research exposed the facial recognition disparities mentioned above. The AJL's mission is to raise awareness about the impacts of AI, equip advocates with empirical research, build the voice and choice of the most impacted communities, and galvanize researchers, policymakers, and industry practitioners to mitigate AI harms and biases. Algorithmic Justice League Their work is where art, advocacy, and research meet. https://www.ajl.org/
The ACLU's Racial Justice Program — This program engages in litigation and advocacy to challenge AI's power to preserve and exacerbate systemic racism and other inequities, working toward building more equitable systems particularly in the areas of employment, housing, and credit. https://www.aclu.org/issues/racial-justice/accountability-in-artificial-intelligence
UN Women — Working at the intersection of gender and AI globally. They are documenting how AI could help governments assess the potential gender impacts of proposed laws, and supporting the development of AI-powered tools that detect digital abuse and connect survivors of gender-based violence to legal services and support. https://www.unwomen.org/en/news-stories/interview/2025/02/how-ai-reinforces-gender-bias-and-what-we-can-do-about-it
Data & Society Research Institute — An independent nonprofit examining the social implications of data-centric technologies, producing research that informs both policy and public understanding of algorithmic harm. https://datasociety.net/
The AI Now Institute — Based at NYU, they focus on the social implications of AI and have been instrumental in influencing regulation around automated decision-making in public systems. https://ainowinstitute.org/
We challenge & reimagine the current trajectory for AI. - The AI Now Insitute's mission
The Bigger Picture
The goal isn't to slow down AI - it's to make sure the people with the most to lose from a biased system have the most say in how it's built. AI has genuine potential to reduce human bias if built deliberately and accountably: standardizing job evaluation criteria, flagging inconsistency in lending decisions, identifying gaps in medical research. The technology can go either way. That's not a reason for despair - it's a reason to get involved.
The future of AI isn't being written by algorithms. It's being written by the people who build them, fund them, regulate them, and demand better from them. That includes you.
Comments