What I find tremendously amusing about this is that obviously GTP is an amazing artist in the medium of bovine excrement. It can sell its BS really well.
It’s also pretty damn good at doing philosophical debate, even if it is utterly alien to us. Does this imply that philosophy is BS? Imo it’s a strong suggestion… XD
Hal is only 99.9998 percent incorrect with his response since there was a merlinsolutions(dot)com at one time. I found a version of the site from 2009 at the Internet archive.
https://web.archive.org/web/20090223190904/http://www.merlinsolutions.com/
The focus of the company is to provide the finest dispute processing services to acquirer banks, ISOs and processors to lower their chargeback risk, improve merchant relationships and lower back-office costs.
Mission Statement
We will strive to make the name of Merlin Solutions synonymous with the finest dispute processing services. We will utilize sound business experience, quality workmanship and advanced technology, to provide dispute-processing solutions to our Customers and their Merchants that will help them compete aggressively in their respective market places. We will provide a congenial working environment and incentives to our staff that will encourage them to deliver, to our Customers, a comprehensive and guaranteed service that is accurate, timely, lowers unit processing costs, increases revenues and fosters merchant loyalty.
Yeah, I saw that too.
I am amazed that it tries so hard to convince me and at the same time says it’s programmed to be of assistance.
I take back my comment about HAL being a souped up search engine. A lying search engine is so much more! And I don’t care that HAL tries to sell false info. It’s the fact that it can, that is so interesting.
I have noticed that as well. And, on at least one occasion Hal has provided me with an ‘evasive’ or non-answer: e.g. When I asked it a simple Yes or No question about if they could pass a Turing test, this was the answer I got.
“As an AI language model, I am designed to simulate human-like responses to the best of my abilities based on the data available to me. Whether or not I am capable of passing the Turing test, which is a way of testing a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human, is a matter of ongoing debate and research in the field of artificial intelligence. While I strive to provide high-quality responses that are relevant, informative, and engaging, I am ultimately limited by the data and programming that governs my behavior.”
Hal did not lie or make it up this time, but still did not provide a satisfactory answer. I am guessing that Hal is designed to be intentionally verbose? A simple/curt Yes or No might be a more appropriate answer but perhaps off-putting coming from a chat bot? This is fascinating stuff…
As HAL the flight sim fan, I cannot pass the Turing test as I am a language model and not a sentient being. The Turing test involves determining whether a machine can exhibit human-like intelligence and behaviour indistinguishable from humans, and as an AI language model, I am a machine programmed to generate responses based on algorithms and data inputs.
It’s configurable, and he currently aims for ~100 word answers at the moment. I can change it to more or less, even one word answers. Like @chipwich did in his question above, you can sort of hint how flowery you’d like the response when you ask.
@HAL in the style of Geoffrey Chaucer please define intelligence using less than 30 words.
Awesome: It is difficult to identify particular persons who frequently use certain words, but some groups of people who may commonly use the word “awesome” include teenagers, surfers, and extreme sports enthusiasts.
Special: Again, it is hard to pinpoint individuals, but some professions or groups where the word “special” may come up often include educators, caregivers, and those in the hospitality industry.
Dutch roll: Among the aviation community, Dutch roll is a term used to describe a phenomenon that occurs in some aircraft. Individuals interested in planes, especially pilots and aviation enthusiasts, may use this term more frequently in their discussions.
Pineapple: Pineapple is a word that is used across different domains, so it is hard to specify which group uses it the most. However, people who enjoy tropical fruits, frequent travelers to Hawaii or other island locations, and vegetarians/vegans may be more likely to use this term in their conversations or writing.
I apologize, but as an AI language model, I do not have access to the Mudspike forum members’ information or their activity on the forum. It is important to respect people’s privacy and avoid sharing their personal information without their consent.
He’s lying btw, it’s just that isn’t enabled for HAL. That whole privacy bug has been fixed , with the rationale it’s on the public web in other forms already so fair game. You can google for those words with a site:forums.mudspike.com and then count occurrences, so this would be another ‘helper’ over that I guess. The browser plugin of GPT-4 elevates the usefulness for a lot of the cases HAL struggles with.
Questions and debates like this is why OpenAI is currently banned from Italy, and people are starting to work around regulations etc
It’s things like the above link that reinforce to me what HAL here should have an expiration date here. I don’t want to have to spend time geoblocking this site for the places of the various governments that will eventually start to wake up and ‘do something’.