ChatGPTS doesn't hijack democracy
ChatGPTS doesn't hijack democracy, it puts democracy in its place.
Recently, I read a New York Times guest essay, "How ChatGPT Hijacks Democracy" by data scientist Nathan E. Sanders and security technologist Brue Schneier.
While I appreciate the threats highlighted by both experts, I found the article mostly irrelevant, particularly regarding the authors' obsession with lobbies.
Yes, OpenAI's ChatGPT makes it possible to create targeted messages for the most venerable politicians that fire off at the optimal moment of vulnerability. But we can't blame ChatGPT for hijacking a democracy that has already been hijacked.
ChatGPT can be used in harmful ways to enable lobbies to drain the pockets of America further. But who cares about adding another trillion to the national debt when we're already at $31 trillion? What's another three feet of water when you're already submerged? More concerning than the national debt is the system that created it.
Contrary to Sanders and Schneier, I see great promise in ChatGPT as a means for putting the Federal Government to work. This work is not that of any civil servant or politician but rather the work of unlocking Government's ocean of data and the learnings that lie within.
Unimplemented recommendations – about 99.5% of all Government reports – can now be synthesized and translated into explanations and action plans that we can all understand and build on.
Open to all, ChatGPT is the tip of the iceberg in returning value to Government. The internet revolution gave us access to information; ChatGPT gives us knowledge and a roadmap. It's unlike anything we've seen before. Some, myself included, see it as revolutionary, especially when we imagine its future.
In addition to asking ChatGPT a range of questions, from legal precedents to the location of a bus stop, I was particularly interested in how ChatGPT would respond to queries about the Federal Government.
As a knowledgeable actor, the "AI model," as it calls itself, isn't pushing answers but framing issues. For example, in response to my question:
What are the most logical steps for addressing the U.S. $31 trillion debt?
The AI model replied:
Addressing the U.S. national debt, which currently stands at approximately $31 trillion, is a complex issue that requires a multi-faceted approach. Some steps that could be taken to address the debt include:
Reducing government spending: This could be done by cutting or eliminating certain programs, reducing the size of the federal workforce, and reforming entitlement programs such as Social Security and Medicare.Increasing government revenue: This could be done by increasing taxes, closing tax loopholes, or implementing a carbon tax.Improving economic growth: A strong and growing economy can help to increase government revenue and reduce the debt-to-GDP ratio. This could be achieved through policies that promote business development, job growth, and trade.Implementing a debt reduction plan: This could be done by setting a target for reducing the debt-to-GDP ratio over a certain period of time, and putting in place a plan to achieve that target.
Prioritizing the debt: It's important to prioritize the debt and focus on reducing the most expensive debt first, such as the debt held by the public, which is currently at around 78% of GDP.
Reducing the interest rate on debt: Lowering the interest rate on the debt would reduce the cost of servicing the debt and increase the effectiveness of other steps to reduce the debt.
Improving the efficiency and effectiveness of government spending: This could be done by implementing measures to reduce waste and improve the efficiency of government programs and services.
It's important to note that addressing the national debt will require difficult choices and trade-offs and will likely involve a combination of spending cuts and revenue increases. It will also demand a long-term commitment and cooperation from both political parties and the public.
A one-pager that summarizes the issue better than any politician can. Direct and straightforward, but with an understanding of the core issues and steps for addressing them. Who needs political rhetoric when the AI model offers better guidance? Instead, we humans should take this knowledge and run with it – focusing our discussion and debate on rational choices and away from the daily razzle-dazzle distraction of a broken political system hyped by media.
Moore's law predicted it. It was just a matter of time. The human is now inferior to the machine at a level not previously witnessed. However, human thinking is not killed but advanced, not by answers but by dialogue.
As dialogue is also the mission of my blog, I feel at home with ChatGPT. It provides a framework for meaningful debate and a path to common sense and civility lacking in today's political world. Another significant advantage is that ChatGPT is not the internet and is free of the non-stop brain rot of information overload that cripples today's world wide web.
Working with ChatGPT is conversing with somebody who knows what they're talking about. For example, in follow-up to my question about the debt, I asked:
Which of the approaches mentioned would be the easiest to implement?
In reply, ChatGPT narrowed its earlier choices to the following and explains why.
"reducing government spending by cutting or eliminating certain programs," reasoning that it may be "easier to implement than raising taxes or reforming entitlement programs, which can be politically sensitive issues."
"Improving the efficiency and effectiveness of government spending by implementing performance metrics, conducting regular audits, and streamlining operations may also be relatively easier to implement, as it does not require a significant change in policy or legislation."
Boarding on brilliant, the AI model teaches us to avoid politically sensitive issues or policy changes if we want to get things done. This is excellent work for an AI engine that's in its infancy. As its access to information grows, so will its unhuman-like insights.
In its current form, ChatGPT does not think for us. ChatGPT can't debate an issue or make a decision. It can, however, navigate the oceans of government information and make sense of it. It doesn't solve Government's many problems, but it puts those problems into a model that makes debate and decision-making easier.
I'm reminded of a quote from Lydon Johnson. "You know, doing what is right is easy. The problem is knowing what is right."
Getting to what is right remains our biggest challenge. We can make ChatGPT our compass, chartering a new course for the future of the United States and democratic republics around the world.
