If you’ve ever wondered how those delightful word salads emerge from the lips of Kamala Harris, wonder no more. After running an experiment with artificial intelligence (AI), I’ve surmised that the Harris campaign might be employing the latest in Large Language Models (LLMs) to generate talking points on everything from immigration to climate. The first thing to know about LLMs, like ChatGPT, is that they harbor an uncanny knowledge of the frequency of word and phrase usage in the English language. Armed with this knowledge, ChatGPT can produce a reasonable continuation of whatever text it’s got so far—by “reasonable” I mean what one might expect someone to write after seeing what people have written on billions of pages in digitized books and articles. It’s a wonderfully built machine for generating trite content, and it’s entirely believable that Harris believes this will rescue her campaign.

Let’s see how the Harris campaign might generate a talking point on say climate. It could start by asking ChatGPT to rank a list of words that might follow the word “climate” at the beginning of a sentence together with their probabilities. ChatGPT perkily responds:

Sure! Here’s a ranked list of words that might follow ‘climate’ at the beginning of a sentence, along with estimated probabilities: change - 30%, action - 15%, crisis - 12%, policy - 10%, impact - 8%.

You continue by asking ChatGPT to rank a list of words that might follow the words “climate change” with their probabilities, and back comes the following:

Here’s a ranked list of words that might follow ‘climate change,’ along with estimated probabilities: impacts - 25%, action - 20%, policy - 15%, adaptation - 12%, mitigation - 10%.

You could continue along this path, stringing together additional words, until ChatGPT generates the following jewel:

Climate change impacts on human health outcomes include respiratory illnesses such as asthma exacerbation due to air pollution from vehicle emissions and exposure to pollutants that can lead to hospitalizations and increased healthcare costs for patients and families.

It makes no sense, but it’s the sort of impassioned plea, devoid of sense, that will resonate with her voters.

Interestingly, the “educated” elites that Harris panders to tend not to see through the nonsense that she presents. What is it about our elites that makes them so vulnerable? Richard Lindzen summarized it best in a lecture to the Global Warming Policy Foundation in 2018 by stating:

1. They have been educated in a system where success has been predicated on their ability to please their professors. In other words, they have been conditioned to rationalize anything. 
2. While they are vulnerable to false narratives, they are far less economically vulnerable than are ordinary people. They believe themselves wealthy enough to withstand the economic pain of the proposed policies, and they are clever enough to often benefit from them. 
3. The narrative is trivial enough for the elite to finally think that they ‘understand’ science. 
4. For many, the need to be regarded as intelligent causes them to fear that opposing anything claimed to be ‘scientific’ might lead to their being regarded as ignorant, and this fear overwhelms any ideological commitment to liberty that they might have. 

There you have it. The makings of what the Democrats believe will be a winning campaign and who knows, they may be right.

Image: YouTube video screen grab, edited.