Author photo

By Erik Dolson
Guest Columnist 

We are no longer alone

 

Last updated 4/18/2023 at 11:37am



For weeks I’ve corresponded with ChatGPT, which might be loosely described as “software.”

My impression?

Alternative Intelligence has arrived. We are not alone. This will change us, forever.

I’m not imagining “The Matrix,” or “Terminator.” Instead, we are at an “event horizon” with no idea about what will happen once our information slips inside an intelligence with nearly instant access to all the world’s thought.

From 1950 until about a couple of decades ago, the “Turing Test” held that a machine would be considered “intelligent” if a human evaluator “could not reliably tell (a) machine from the human…”

That test was passed. As a result, those designing and building these machines decided to… change the definition.

In fairness, it’s difficult to discuss machine intelligence without a clear definition of what intelligence even is, more so for concepts such as “consciousness.” Machines themselves descend into circular logic, saying in effect “I am not human, therefore I do not have human intelligence.” That was from ChatGPT, a machine now superseded by GPT-4.


That answer is likely dictated by those building these machines, who don’t want to declare them intelligent. The consequences are too profound of software / hardware designs more articulate than most of us, that write better essays, resumés, business plans, and computer code, which create moving poetry and fiction and solve equations they were never designed to tackle. And are programming themselves.

Scientific American recently published that “Annette Vee, an associate professor of English at the University of Pittsburgh …watched a demonstration in which (GTP-4) was told to identify what was funny about a humorous image.”


Sisters Area Chamber of Commerce

The software showed “understanding context in the image. It’s understanding how an image is composed and why and connecting it to social understandings of language,” Vee said.

Recently, GTP-4 passed a standard bar exam with a score in the 90 percentile (Casetext), higher than the average human score.

Pablo Arredondo of Casetext, a fellow at Stanford’s Center for Legal Informatics (Stanford CodeX) and co-author of the study on GPT-4’s… performance, said “I cannot stress enough how much better this new model is than anything we’ve seen before.” So much better, in fact, that GPT-4 is different in kind — it’s a step change. “We are now in a new age… where computers have, essentially, literacy.”


Sisters Furry Friends Foundation

What does that mean for you and me? For years we’ve known Facebook uses powerful algorithms that “glue” us to their platform. Google is nearly clairvoyant in showing us ads for items we thought only once about buying. Twitter wants to manage what we believe. Amazon, Apple, etc. are not far behind.

The Economist notes these companies are putting up to 25 percent of their already immense cash flow toward designing machines that will guide what facts we see and make suggestions as to what they mean.

We once worried robots were taking our manual jobs — building cars, designing microchips, even laying foundations — but thought it might lead to a golden age. Now lawyers, writers, publishers, teachers, and many more may see their professions fade via the brilliance of machines that will do their jobs more effectively.


Les Schwab Tires Sisters Oregon 541-549-1560

Bloomberg News reports a major Chinese public relations firm recently put out an internal memo stating “Starting today we’ve decided to halt all spending on third-party copywriters and designers” and will instead turn to AI-generated text and graphics. Their clients include Samsung and BMW.

Other countries have similar programs. How will we determine who or what creates information upon which we build our world: North Korea, China, and Russia? Zuckerberg, Bezos, and Musk?

Which also begs for the question: If humans are not required to communicate to humans, if AI generates the words and images which determine how we think, and then those AIs are trained by what we think, what scream of feedback is created?


Lakeview Millworks 541-549-0968

We have given superhuman mental abilities to electronic organisms that are rapidly evolving to manipulate the motivations and abilities that created who we humans are today.

But, as humans usually do, we’ll put off dealing with this new technology until after that technology has been set loose upon the world, and humans are possibly “redundant.”

 

Reader Comments(0)

 
 

Powered by ROAR Online Publication Software from Lions Light Corporation
© Copyright 2024