Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
All this media fascination with AI (enough apparently to create an AI thread on C-D's Great Debates) makes me yearn for a bumper sticker or tee saying, "AI is the A[nti]C[hrist]."
Fear of the unknown
Instead of fearing "Artificial Intelligence", we should fear the "Organic Stupidity" in the White House. This is a far more dangerous, immediate threat to the nation.
My guess is MSM has been tasked to start preparing the public for AI that will soon be showing up soon. One task of big media (MSM, movies, tv, etc) which is owned by a few big corps, is to prepare the public for things that big corps want to do.
Right now we are hearing about chatgpt which is a text response. But big corps are probably ready to implement AI in all kinds of services. They want the public to accept AI when you deal with a company. The fed gov already has an AI office to promote AI in gov and business.
All this media fascination with AI (enough apparently to create an AI thread on C-D's Great Debates) makes me yearn for a bumper sticker or tee saying, "AI is the A[nti]C[hrist]."
Orwell's book 1984 introduced us to the concept of Thought Police.
Now you have it.
If the AI is programmed to keep you from writing "bad: words or thoughts, then you won't be able to write them, right?
I take it you haven't used MS Word in the last 30 years.
You don't see all those blue underlines and dotted underlines. That Microsux AI telling you to write something the way Microsux wants you to write. Not to long from now, you won't have a choice. You'll have to do it the way Microsux wants you to do it.
Here is the true danger of Artificial Intelligence, and am speaking as someone who is in semicon manufacturing: making people stupid.
Am not speaking facetiously. At best, only 1/4 of what I said is a joke.
At the theoretical level, AI is essentially a search engine no different than that of Google, Bing, or Duckduck Go. But at the application level, this is where ChatGPT et al are vastly different.
When you do a search on 'health and fitness', for example, what the search engines will do is trawl the internet for 'health and fitness' websites based on certain criteria, and one of the criteria is popularity, then the search engine, Google for example, will display those websites. Then it is up to you to go thru each websites and filter out the information you want/need. How much and how long you read thru the websites is up to your endurance and most important: INTELLIGENCE. Basically, the smarter you are, the better at filtering out the info for what you want/need.
What AI does is incorporate 'large language model' (LLM) capability along with its own search capability to produce near human level responses. The LLM is essentially the entire English language which includes the dictionary and rules of grammar.
Now the AI will go thru the most relevant 'health and fitness' websites that you asked for, and produce detailed questions for you, then as you replied to those questions, the AI will filter those websites to fit your responses.
Here is MY question to ChatGPT on comic books characters...
Q: Is the One Punch Man more powerful than Superman?
A: There is no straightforward answer to this question because both One Punch Man and Superman exist in different fictional universes and their powers are defined by the rules and limitations of their respective universes.
In the world of One Punch Man, the protagonist Saitama has the ability to defeat any opponent with a single punch, regardless of their size, strength, or other abilities. However, the extent and limits of his powers are not fully explored, and it is unclear how he would fare against opponents with reality-warping abilities or other types of unconventional powers.
In the DC Comics universe, Superman is one of the most powerful superheroes, with a wide range of abilities including super strength, speed, flight, invulnerability, and heat vision. He has also faced a variety of powerful opponents, both physical and metaphysical, and has managed to overcome them with his skills and powers.
Ultimately, it is difficult to compare the two characters because they exist in different fictional universes and have different sets of powers and limitations. It is also worth noting that the power levels of fictional characters can vary widely depending on the specific storyline or interpretation, so any comparison is subjective and open to interpretation.
Now, here is a more scientific and technical question that has national security implications...
Q: In designing a low radar observable platform, what is the ten lambda rule and what shapes are applicable?
A: The "ten lambda rule" is a principle used in the design of low radar observable platforms, which states that the radar cross section (RCS) of an object will be reduced if its dimensions are smaller than ten times the wavelength of the radar signal being used. In other words, if an object's dimensions are less than ten times the wavelength, the radar waves will be scattered in different directions and will not return to the radar receiver, making the object less detectable.
Regarding the shapes that are applicable in the design of low radar observable platforms, there are several geometries that can be used to reduce the radar signature of an object. Some of the shapes that have been found to be effective include:
Faceted shapes: These are objects that have flat surfaces at different angles to one another, creating multiple surfaces for the radar waves to bounce off in different directions.
Rounded shapes: Objects with smooth, curved surfaces can help to scatter radar waves in different directions, making it more difficult for the radar to detect them.
Angled shapes: Objects that are designed with angled edges and surfaces can help to deflect radar waves away from the receiver, reducing the RCS of the object.
Blended shapes: These are objects that combine different shapes and geometries to reduce their radar signature. For example, a blended shape might have both faceted and rounded surfaces to scatter radar waves in different directions.
It's worth noting that the effectiveness of these shapes can vary depending on the specific radar frequency and detection method being used, so a combination of different shapes and materials may be necessary to achieve the desired level of low observability.
See the differences and the commonalities? For each question, ChatGPT gave the response in less than one second.
For the question about comic books characters, you can get that answer at any comic-con or neighborhood comic books stores.
But 9 out of 10 people will not know of the 10 lambda rule and how it is applicable to designing a 'stealth' airplane. You would have to finish at least a BS then work for Lockheed or Boeing for yrs in order to appreciate the answer that ChatGPT gave me. The 10 lambda rule is minimally applicable to the F-117 but is the foundation for the F-22, F-35, B-2, B-21, and the NGAD fighter. I understood the answer because I already knew the principles involved based on my understanding of basic radar detection and my USAF time in service.
This level of interaction is what alarmed many technologists, including some who helped created the AI technology, about the dangers of AI. The responses, even benign like about comic books, depends on WHO DOES THE AI PROGRAMMING IN THE FIRST PLACE. The near human level quality will seduce people into believing that what they received is all that there is. No different than what social media companies are doing now.
Artificial Intelligence WILL be the equivalent of the steam engine that changed transportation, or the printing press that changed global politics, or the machine gun that changed warfare. Each change is a point of no return, meaning that if we regress to before the change, we go back at least one hundred yrs.
The problem here is that if you own a machine shop, you are 1/2 way to making your own steam engine, or printing press, or machine gun. But AI programming is opaque. Each AI engine is unique to its creator(s) and their intentions.
At the theoretical level, AI is essentially a search engine no different than that of Google, Bing, or Duckduck Go. But at the application level, this is where ChatGPT et al are vastly different.
Maybe I'm just ignorant or maybe the technology is too new, but my limited exposure to AI has been underwhelming. I don't see AI replacing white collar jobs, as noted further upstream, but I see it as another tool in the toolkit.
It appears to me that, in its current iteration, AI is simply converting data to sentences. It's not "intelligence" to the extent that it can create something from nothing, rather it creates something from something. My son played high school baseball 8 years ago and the Game Changer app converted the scorecard into a game recap narrative. It took data and converted it to sentences. Cool but not telling me anything I couldn't gather from looking at the scorecard.
One hobby that sucks up a ton of my personal time is ancestry. In particular, finding someone who immigrated here at the turn of the 20th century, who I cannot find records for. I thought maybe I'll try this just for s&g. I asked Bard "What can you tell me about [person's name] who immigrated here in 1906 and married [wife's name]?" I refined the question with specific information but came to the conclusion that AI is simply taking all the links that would appear on a Google search, aggregating them and generating a story. A story that was right in some areas, and incorrect in others. One in particular gave information that was fictitious. I researched the people, places and things mentioned and some were never in existence.
Long story short it's a tool or a jumping off point, if you will, but it's not a replacement for actual research and logic. At least not currently.
...but my limited exposure to AI has been underwhelming.
So do most of us. I work in semicon manufacturing. My company produces memory for everything from 'smart' refrigerators to personal computers to -- yes -- Artificial Intelligence computers. And even so, it took me a few months of personal research and talking to some people at work to have just a high level understanding of AI technology.
Quote:
Originally Posted by BoSox 15
I don't see AI replacing white collar jobs, as noted further upstream, but I see it as another tool in the toolkit.
It appears to me that, in its current iteration, AI is simply converting data to sentences. It's not "intelligence" to the extent that it can create something from nothing, rather it creates something from something. My son played high school baseball 8 years ago and the Game Changer app converted the scorecard into a game recap narrative. It took data and converted it to sentences. Cool but not telling me anything I couldn't gather from looking at the scorecard.
Note the highlighted. What you said correlates exactly to my post.
The reason you could easily dismissed the Game Changer app is because you ALREADY knew the sports data. Same as how I understood ChatGPT responses about the 10 lambda rule and 'stealth' fighter designs because I ALREADY knew about radar detection principles. In both situations, AI is another tool that we can use. But for those who are not versed in our respective knowledge, the Game Changer app and ChatGPT responses will be the only tool they know. Which is a benefit and which can be a danger?
I take it you haven't used MS Word in the last 30 years.
Oh yeah I have. A gazillion times.
You don't see all those blue underlines and dotted underlines. That Microsux AI telling you to write something the way Microsux wants you to write.
Yes I do see them, not just on Word but sometimes while typing on mb's like C-D. I agree that is an example AI at work, I just have my pants up my you know what about the media saturation.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
