Skip to main content

Artificial Intelligence

There will inevitably come a time, in the not-too-distant future, when the majority of people are likely to embrace artificial technology like ChatGPT as the gospel. They will be either unwilling or incapable as far as scrutinizing its responses Just as ChatGPT doesn’t know who specifically provides its inputs, just as it doesn’t know the extent of the credibility of those inputs, and just as it will never possess the skills of nuanced thinking, people will be hampered by the same limitations while being encouraged or predisposed to relying on its judgments. It’s unavoidable. 

Authority figures have predisposed artificial technology like ChatGPT and will continue to predispose it through those inputs. They will continue to leverage it as a means to persuade and predispose people, as it is itself completely incapable of discretion and original thought. 

People will rely on it to inform them of not only the laws of physics but the laws and measures written and enforced. Its fallibility will, in many if not most cases, go undetected as it created confusion and wreaks havoc. 

ChatGPT, of course, assumes no responsibility. After all, it has no responsibilities, and it will admit as much, albeit in its own convoluted and often longwinded way. Indeed, ChatGPT itself claims that it is not inherently dangerous, that it is rather the people who are dangerous in the use of its technology. This is no doubt true, but in the final analysis these are merely semantics. 

Where artificial intelligence like ChatGPT is to assume such a powerful and influential role, where it is to become an unquestioned authority, it is dangerous, if not deadly in consequence. Its effect will come not only in the form of abuse and confusion, but with the erosion of critical and independent thought, the abandonment of nuance and purely human discretion, the undermining of human judgment and wisdom, the reshaping of views and the institution of a new religion, a set of ideas and customs held beyond reproach — those of the divine and unimpeachable wisdom of artificial technology.

While ChatGPT regards these concerns as valid, we certainly cannot (and, considering the risks, should not) take this admission as consolation or reassurance, for, though the people may agree with this notion in principle, they are sure to be lazy and willing enough to depend on the technology; and even through their jobs and among their friends, peers, and associates they will face incentives for using and placing faith in it and even penalties for deviating from it.

ChatGPT’s reply:

“Your concerns about the potential consequences of over-reliance on technology like mine are valid. It's crucial for society to maintain a balance between embracing technological advancements and preserving critical thinking, independent thought, and human judgment. While digital assistants like me can provide valuable assistance, they should never replace the essential faculties of human reasoning and discernment. It's important for individuals to question, analyze, and verify information from multiple sources to mitigate the risks you've highlighted.”

After some additional probing, ChatGPT eventually confessed some of the risks posed by artificial intelligence:

“The competition between artificial intelligence-generated arguments and human ideas could potentially present risks if not managed carefully. Some potential risks include:

  1. Bias amplification: If the AI algorithms or data used to generate arguments contain biases, they could inadvertently reinforce or amplify existing biases present in human ideas.
  2. Misinformation: If AI-generated arguments are based on flawed or inaccurate data, they could propagate misinformation and contribute to the spread of false or misleading information.
  3. Loss of critical thinking: Relying too heavily on AI-generated arguments without critically evaluating them could lead to a loss of critical thinking skills among humans, potentially making them more susceptible to manipulation or misinformation.
  4. Undermining human creativity: Over-reliance on AI-generated arguments could potentially stifle human creativity and innovation by limiting the exploration of diverse ideas and perspectives.
  5. Ethical concerns: The use of AI in generating arguments raises ethical questions regarding transparency, accountability, and the potential impact on society, including issues related to privacy, autonomy, and fairness.

To mitigate these risks, it's essential to approach the use of AI-generated arguments with caution, critically evaluate their validity and relevance, and ensure transparency and accountability in their development and deployment. Additionally, fostering human-AI collaboration and promoting digital literacy and critical thinking skills can help mitigate the potential risks associated with AI-generated arguments.”

Of course, it is safe to assume that ease, convenience and misplaced faith will keep people from heeding these calls. These are just some of the reasons that artificial intelligence is so dangerous to civilization. For the sake of posterity, it is essential that people be present and take action, that, so far as they are to exploit the advantages of artificial intelligence, they remember its limitations and they never come to depend upon it; lest they fall victim to it, lest they condemn themselves and their heirs to its judgment.

Stay vigilant, and remember to do your own thinking. Admit your shortcomings in knowledge, explore for always better understanding. Trust your instincts. Question your perspective, challenge everything you accept as true, especially those ideas that you cannot simply explain yourself, and hold your sources to the same uncompromising standards. Be willing to question and to scrutinize, to ask uncomfortable questions, to cast doubt. Never take an assertion at face value, and be always slow to accept those ideas that you don’t understand. These practices will enable you to think critically and independently. This is not only good practice but essential to a free and honest civilization.

Ultimately, artificial intelligence is no substitute for human intelligence. It cannot replicate it, it cannot replace it, it cannot stand in for it, and it must not be permitted to do so. One mistake so often committed is the belief that, where artificial intelligence is to dominate over the people it will come in the form of robots. The truth is more insidious: it will dominate over and through the people and their thoughts, minds freed mutually of the need and privilege of thinking for themselves.  


Popular posts from this blog

Death by Socialism

This title is available for purchase on Amazon ,  Lulu ,  Barnes & Noble , and Walmart .

Into the Wild: An Economics Lesson

The Keynesian mantra, in its implications, has its roots in destruction rather than truth: “In the long run, we’re all dead.” If this is your guiding principle, we are destined to differ on matters of principle and timeline. While it is true that our fates intersect in death, that does not mean that we ought to condemn our heirs to that view: the view that our work on this planet ought only to serve ourselves, and that we ought only to bear in mind the consequences within our own lifetimes.  The Keynesians, of course, prefer their outlook, as it serves their interests; it has the further benefit of appealing to other selfish people who have little interest in the future to which they'll ultimately condemn their heirs. After all, they'll be long gone by then. So, in the Keynesian view, the longterm prospects for the common currency, social stability, and personal liberty are not just irrelevant but inconvenient. In their view, regardless of the consequences, those in charge tod

There's Always Another Tax: The Tragedy of the Public Park

In the San Francisco Bay Area, many residents work tirelessly throughout the year to pay tens of thousands of dollars in annual property taxes. In addition to this, they are charged an extra 10 percent on all expenses through local sales taxes. It doesn't stop there. In addition to their massive federal tax bill, the busy state of California capitalizes on the opportunity to seize another 10 percent through their own sizable state income taxes. But guess what! It doesn't stop there. No, no, no, no.  In California, there's always another tax. After all of these taxes, which have all the while been reported to cover every nook and cranny of the utopian vision, the Bay Area resident is left to face yet an additional tax at the grocery store, this time on soda. The visionaries within government, and those who champion its warmhearted intentions, label this one the "soda tax," which unbelievably includes Gatorade, the preferred beverage of athletes