24
Replies
Just saw an article about Buzzfeed. They're shutting down their news division, but trying to make up losses by using AI to write articles.
By @Chuck Uebele
===========
AI articles can't be protected by copyright.
For more AI developments, watch this interesting video (23 min).
AI articles can't be protected by copyright.
By @Nancy OShea
...in the USA. My AI articles are (c) protected, also in the USA. I just need to prove the creation of a reasonable sophisticated prompt. "Write a letter to Nancy" would not be enough:
Dear Nancy,
I am writing to you to express my gratitude for your support and guidance during my time at ABC Inc. You have been a great mentor and friend to me, and I have learned a lot from you. I appreciate your kindness, patience, and professionalism. You have always encouraged me to pursue my goals and challenged me to grow as a person and as an employee.
I am sad to leave ABC Inc., but I am also excited for the new opportunity that awaits me at XYZ Ltd. I hope we can stay in touch and continue to share our experiences and insights. You can reach me at my personal email address or phone number anytime.
Thank you again for everything you have done for me. I wish you all the best in your future endeavors.
Sincerely,
Your name
Like I said in a previous post, one of the AIs accused someone of being a murderer when they queried it about an incident which happened some years ago.
By @osgood_
==========
I must have missed that post. Misinformation is highly subjective. It's based on one's perception and available data. In the absence of reliable information, confabulation occurs. Flat-Earth theorists firmly believe the world isn't round. All AI machines hallucinate, an odd quirk for which programmers have no explanation.
Imagine a self-learning machine gives a potent breakthrough formula to a scientist.
- Scientist A uses the formula to create a weapon that can kill millions of people.
- Scientist B uses the formula to create a cure for cancer that can save millions of people.
Based on available information, who is the saint & who is the sinner?
Now let's further imagine that
- Scientist A works for the military in a war-torn region like Ukraine. The weapon is never deployed but acts as a deterrent which leads to a quicker end to the war. A peace treaty is signed resulting in saved lives on both sides.
- Scientist B works for a pharmaceutical company that fears releasing the cancer cure now will adversely impact profits from their other drug treatments. The cancer cure is withheld resulting in lives lost to cancer.
Now who are the saints & who are the sinners?
The more information we have to draw from, the more likely we are to arrive at better conclusions. It's no different for AI.
Insofar as AI being good or evil, it's neither. It's merely a tool, no more good or evil than a 3D printer. AFAIK, nobody has successfully sued a 3D printer (yet).
Therefore an hallucinating AI or human with mental illness can't be sued for things they say as they are not of sound mind.
Is the company behind an hallucinating or mentally ill robot not responsible for its well-being and therefore should they not be responsible if they choose to let their creation loose on the public?
I really dont have much time for AI as in many cases its proven to be 'not fit for purpose' and unfortunately, what is even more scary, neither are the majority of people that may use it and hang on its every word. Put both together and its pretty concerning of what could be said and suggested.
AI's ultimate success & impact on society will depend on how humans use it & regulate it.
By @Nancy OShea
That's what troubles me, l have no faith in the human race, it abuses everything and anything it can.
I think the one positive, from what l've seen at least, you have to know what to ask an AI robot to get it to provide you with anything usable, particularly in reference to coding, which requires some knowledge, at which point you probably know what you're doing anyway, so l just view it as another walking stick used by those who aren't professional and are just hacking things together.
It's still a concern that there is a very real possibility that it becomes a source for the lemmings out there though, of which there are more, than not, who will use it to make a quick buck regardless of if the information is correct or not.
Social media has proven to be a bad idea since its conception in my view, as has youtube to a certain extent, who are doing their best to ban/suspend those channels which post mis-information from crackpots who say they are being denied free speech but are really just in it for maximum views and the revenue.
crackpots who say they are being denied free speech but are really just in it for maximum views and the revenue.
By @osgood_
=========
Sadly, you've just described our U.S. House of Congress. The more outrageous things they say, the more TikTok views they get. It's a sickness.
If I have fears about AI it's that it might have been trained on social media garbage.
That's kind of my point. Will actual humans fact check things? They're not doing the best job now.
By @Chuck Uebele
You've hit the nail on the head. The more widely available these kinds of solutions become to the general public the more potential for incorrect/mis information to be spread. Let's face it the majority of humans are generally lazy no good for nothing imbeciles, just out to make money, have no pride and even less morales. Many will just take what an AI returns at face value, without going through rigorous checks.
AdChoices