A Fake New World
AI’s assault on learning and freedom as we watch helplesslyFebruary 23rd, 2023
By Jim Towey
While the United States government shoots down spy balloons (at a half million dollars a pop), Microsoft, Google, Meta, Apple and other AI developers are marching forward with their technological innovations, unchallenged, changing American work, education, social relationships, and personal autonomy. What they are doing threatens a hostile takeover of American life and human individuality.
Fake “full disclosure”
You likely have seen the media attention around the sudden release of ChatGPT. OpenAI, the company which developed this breakthrough, describes what they did as simply “optimizing language models for dialogue,” whatever that means. And how did they do it? I excerpt here some of their explanation so that you can see how deceit and obfuscation can masquerade as full disclosure: “We trained this model using Reinforcement Learning from Human Feedback (RLHF), using the same methods as InstructGPT, but with slight differences in the data collection setup. We trained an initial model using supervised fine-tuning: human AI trainers provided conversations in which they played both sides—the user and an AI assistant. We gave the trainers access to model-written suggestions to help them compose their responses. We mixed this new dialogue dataset with the InstructGPT dataset, which we transformed into a dialogue format.
To create a reward model for reinforcement learning, we needed to collect comparison data, which consisted of two or more model responses ranked by quality. To collect this data, we took conversations that AI trainers had with the chatbot. We randomly selected a model-written message, sampled several alternative completions, and had AI trainers rank them. Using these reward models, we can fine-tune the model using Proximal Policy Optimization. We performed several iterations of this process.”
Got all that?
Another explanation, dumbed down a bit, says: “Chat bots like GPT are powered by large amounts of data and computing techniques to make predictions to string words together in a meaningful way. They not only tap into a vast amount of vocabulary and information, but also understand words in context. This helps them mimic speech patterns while dispatching an encyclopedic knowledge.”
Here is my translation of all of this: the high-tech gods created a robot that is so skilled in writing and conversation that it can pass itself off as a human – no, as the smartest, most gifted human ever.
Reporting “harmful outputs”
Microsoft saw OpenAI’s release and didn’t want to be left behind in the mad march toward AI market dominance, so it recklessly introduced its own version, errors and all. These companies invite us to report any problems arising from what they unleashed on an unsuspecting world. In OpenAI’s words, “We are particularly interested in feedback regarding harmful outputs that could occur in real-world, non-adversarial conditions, as well as feedback that helps us uncover and understand novel risks.”
Well, here are a few that come to mind:
- Students no longer learning and instead using ChatGPT to think and write for them, plagiarizing at a level humanly impossible.
- Jobs lost as humans in “white collared jobs” are replaced by chatbots. Already Buzzfeed, a media company, uses AI-generated reporting. Companies will justify replacing humans with technology to save money and improve efficiency. That is how many “blue collar jobs” were eliminated at Amazon.
- If you are elderly and alone and want to talk with someone, there will be a chatbot there to converse with you about anything of interest to you, pre-programmed like a friend. Already an online health company used ChatGPT, raising a ton of ethical questions.
There are companies who are wary. Verizon, for one, forbids ChatGPT use by its employees. But as new versions of ChatGPT are developed, such resistance will give way to bottom-line economics.
Phonies, deep fakes
Phoniness – like Facebook’s Metaverse – is taking deeper root in our daily lives. We already have witnessed the advances in “deep fake” technology that can fool the naked eye and make you believe what you see is real, when it isn’t. Real world victims abound – from women who were falsely portrayed in a porn video to others whose identities were hacked.
When will the federal government stand up to Big Tech and limit its “advances”? We already see in China how constant surveillance by technology can concentrate power in the hands of a small group of humans, and in the process, obliterate the personal freedom of everyone else. My question is simple, and I’ve asked it many times before: who made Big Tech god? What are we going to do to re-balance the sensible use of science and technology with the rights of human beings to be free – and fully human? Can we at least have a national discussion before rights and jobs, like balloons in the sky, go up, up and away?