Artificial Intelligence Can Make You Look VERY Unintelligent
Ok I admit it. I am one of those few remaining people that still plays wordle every day.
One evening I was playing the daily game, and I wondered if ChatGPT could do a better job than me.
I gathered the letters I had eliminated from the puzzle and asked ChatGPT to find me possible words.
So my question was along the lines of "Find a 5 letter word that does not include the letters S, L ,I M, E, C, H, A. R. T"
Quick as a flash the AI came back and confidently told me - one word which doesn't include any of these letters is "Giddy".
Hmmm, but giddy has an "i" in it i told the AI - and on went the conversation with ChatGPT continuing to give me examples which still included letters I asked it to exclude, each confidently stating that the example provided was correct. (you can see the screen hot of the conversation below)
Recommended by LinkedIn
The point is this. I had enough knowledge on this question to be able to see if ChatGPT was giving me a correct or incorrect answer.
But what happens when you ask a question on a topic about which you have no personal knowledge?
ChatGPT or other generative AIs can provide an answer, presented in a way which looks compelling and like an unquestionable fact, but the "facts" it provides can be totally wrong, and if you as the questioner don't have the background knowledge yourself, you can end up believing the incorrect information provided.
This is compounded by the fact, that these generative AI's are so good at conversation and writing that they come across as "intelligent" and all knowing. that's where the danger lies.
I have already seen this manifest even with our own journalists. They have relied on a generative AI to produce an article, it comes out well written but "misses" the point and exposes the author as having no knowledge on their subject matter to any reader that is well informed on that topic.
The answer, use generative AI but don't accept what it comes up with, fact check and sanity check for yourself, using the time old method of checking against verifiable sources if you don't you run the risk of ARTIFICIAL INTELLIGENCE making you look like you are REALLY UNINTELLIGENT!
Do better. Be better.
6moBrilliant write up!
COO at The Virtual Connexion ❂Advisor to SMEs ❂ Local and International Consulting ❂ Mentor to C-Suite Leaders ❂ Staffing Requirements ❂ Talent Assessment ❂ Start-ups
10moChatGPT apologises a lot, even after reading "that's not what I want" from me - what a sweetheart!! 😂 😂 😂
Andrew I saw you play pickleball at Bangsar Sports Complex last year, I recognised you, I wanted to say hi..but I felt like a fan who might get brushed off by a celebrity. ..😆 anyway Happy New Year and God bless and great postings as always.
I am like a blacksmith that pursuing to create a legendary weapon through good ores and smithing skills. Except that it is Facilities Management than a weapon.
10moI had a funny incident recently when my lecturer assumed that I submit my commentaries for a scripture we required to study on, as generated by our pal ChatGPT. I was dumbfounded and can only think: So.. Am I a robot? Maybe I am ChatGPT incarnation in a human form?
Fixing data center operations
10moI ran the test using a somewhat different phrasing compared to yours. Brown came back as the answer. Total legit.