Thought on AI

UKworkshop.co.uk

Help Support UKworkshop.co.uk:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
i have found it wrong on many facts , and when corrected , it apologises and then gives a different WRONG answer , sometimes on the 2nd attempt it gets it right , BUT i have had to try 3 times in some instances
Writing articles about , artists , record covers , battle of Britain , inventors , authors, and sport - Cricket particularly "the 100" and also cricket ground stats on T20 top score , and cricketeer who achieved it.
on the cricket it got everything wrong , no matter how many times I tried - - to name just a few examples
this was last year on chatgpt
so do beware if fact needed

I use a lot for organising a few parargraphs on blog / facebook and website pages - for posts - which work well
 
Reading through the answer that ChatGTP gave is interesting. It is certainly logical, but weighted toward the use of a jig. There is no advice that practising ones skills, will allow you to dispense with sharpening jigs - only that some people like to do it that way because of " the flexibility and the skill involved".

As for it's reliance on the internet - the other day, I was looking up information on William Lee, the inventor of the knitting machine. Checking through the results I discovered that most of the sites were regurgitating the same information. And in some cases, it was verbatim. There is a lot of this " copying of others homework", which also includes the mistakes as well. And these "mistakes" can all too quickly become the new "truth" according to the internet. This could well be the reason that ChatGPT can also get it wrong.
 
Reading through the answer that ChatGTP gave is interesting. It is certainly logical, but weighted toward the use of a jig. There is no advice that practising ones skills, will allow you to dispense with sharpening jigs - only that some people like to do it that way because of " the flexibility and the skill involved".

As for it's reliance on the internet - the other day, I was looking up information on William Lee, the inventor of the knitting machine. Checking through the results I discovered that most of the sites were regurgitating the same information. And in some cases, it was verbatim. There is a lot of this " copying of others homework", which also includes the mistakes as well. And these "mistakes" can all too quickly become the new "truth" according to the internet. This could well be the reason that ChatGPT can also get it wrong.
An individual doing a google search independently would find exactly the same set of results and, probably, form precisely the same mistaken conclusion. ChatGPT at least ensures (I assume) that all highly rated digital data is accessed - us humans are frequently less thorough!
 
So much AI hype at work :) Has anyone seen a meaningful project / outcome as a result of AI / GenAI in the past couple of years at their work? We've had lots of talk about it but no-one has actually demonstrated anything yet
 
So much AI hype at work :) Has anyone seen a meaningful project / outcome as a result of AI / GenAI in the past couple of years at their work? We've had lots of talk about it but no-one has actually demonstrated anything yet
Actually the foundation of what I’m involved in at work is AI and the products are in use widely today. Frankly what it does couldn’t be done as reliably any other way.
 
I have read too much Sci fi to be fully comfortable with the ubiquity of AI in our lives.
It is likely unstoppable at this point and already pretty clever in some ways. I am amazed at some of the AI art that is produced. But I also agree with those artists who are not happy their work is being used to train the systems without permission or compensation.
Similarly, every time we use chat gpt we are helping it learn.
Are we teaching our future maters how to best control us ?! 🤣
 
Actually the foundation of what I’m involved in at work is AI and the products are in use widely today. Frankly what it does couldn’t be done as reliably any other way.
Good to hear.. it wasn’t meant as a criticism - more my place of work lacking any ability to execute on anything to do with AI :)
 
Same here, develop lots of in product use of AI & ML. The ML stuff has been used for many years and last few years wrapping AI into it. Using language models in apps connected to digital twins and data lakes, works well, saves time/money etc. Not good at innovation, which is what you'd expect, but time saving to give teams more time to innovate or a direction to innovate to that will be most successful (potentially).

We've benchmarked the general model and it compares well to mid stage career people, so knock out a quick report or ppt with reference material from the internal sanctioned data sets, but you need some mk1 eyeballs in the room to draw conclusions.

Inflated expectations vs reality, gartner hype curve stuff.
 
Same here, develop lots of in product use of AI & ML. The ML stuff has been used for many years and last few years wrapping AI into it. Using language models in apps connected to digital twins and data lakes, works well, saves time/money etc. Not good at innovation, which is what you'd expect, but time saving to give teams more time to innovate or a direction to innovate to that will be most successful (potentially).

We've benchmarked the general model and it compares well to mid stage career people, so knock out a quick report or ppt with reference material from the internal sanctioned data sets, but you need some mk1 eyeballs in the room to draw conclusions.

Inflated expectations vs reality, gartner hype curve stuff.
Sorry, did you just experience a bad case of human verbal diarrhoea ... or did you simply highlight, copy and paste the verbal diarrhoea that, after a prompt of some sort, your preferred AI chatbot thingy spewed out for you? Slainte.

PS. My tongue, may, or may not have, been firmly in my cheek when I typed all the above myself with my own fingers. Any and all typos and careless expressions are therefore mine to claim.
 
Sorry, did you just experience a bad case of human verbal diarrhoea ... or did you simply highlight, copy and paste the verbal diarrhoea that, after a prompt of some sort, your preferred AI chatbot thingy spewed out for you? Slainte.

PS. My tongue, may, or may not have, been firmly in my cheek when I typed all the above myself with my own fingers. Any and all typos and careless expressions are therefore mine to claim.

No sorry Richard, tis the language i use monday to friday and i'm known for my brevity and clarity, welcome to the world of corporate engineering!

I apologies for using it on a weekend.
 
For me the term artificial intelligence is misleading just as much as calling a phone smart or thinking there is anything social about social media.

In what sense are they defining intelligence ? I would suggest that it has to be a lot more than pattern matching or just logical comparisons and that for a minimum you would just state the what and then the intelligence should solve the how.
 
Jacob, you should register with ChatGPT and carry on your sharpening argument with it. Fill your boots and give us all a break. While you're there ask it about the 2019 election.
 
What exactly is wrong with this answer?
It just trots out modern sharpening dogma and is wrong on several points, if not all of them.
Ditto with flattening stones answers.
Maybe you should also ask it for alternative views so you get a choice?
In the long term the benefits could be huge, but only if they are distributed. Going by past history of technological advance this may not happen without a struggle. https://www.theguardian.com/society...ersal-basic-income-about-to-transform-society
 
Last edited:
It just trots out modern sharpening dogma and is wrong on several points, if not all of them.
Ditto with flattening stones answers.
Maybe you should also ask it for alternative views so you get a choice?
Well this is interesting

Are you aware of the sharpening technique espoused by Jacob of UKWorkshop

Jacob's sharpening technique, as discussed on the UKWorkshop forums, emphasizes traditional methods and freehand sharpening over the use of jigs and modern gadgets. He advocates for the use of oilstones and a minimalist approach to equipment, arguing that this method is both cost-effective and efficient when mastered.

Jacob's approach involves the use of a few oilstones to sharpen tools by hand, a technique he believes offers better control and results than using jigs. He often mentions that the over-reliance on gadgets and jigs, which became popular in the DIY boom of the 1980s, has overshadowed these traditional methods. His technique is rooted in consistent practice, developing muscle memory and skill, which he argues is more effective in the long run than using guides or jigs [oai_citation:1,sharpening technique | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening-technique.74627/) [oai_citation:2,sharpening technique | Page 6 | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening-technique.74627/page-6) [oai_citation:3,Sharpening | Page 5 | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening.144505/page-5).

He also stresses the importance of maintaining a flat surface on the sharpening stones, a critical aspect of ensuring a sharp and consistent edge on tools. Jacob often points out that this traditional method, while requiring more initial effort and practice, leads to a deeper understanding of the sharpening process and ultimately more precise results [oai_citation:4,Sharpening | Page 17 | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening.137789/page-17) [oai_citation:5,Sharpening | Page 10 | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening.137789/page-10).

For those interested in exploring his methods, it's advisable to start with a set of good quality oilstones and to practice regularly, focusing on achieving a consistent bevel angle and a keen edge through freehand techniques.
 
Well this is interesting

Are you aware of the sharpening technique espoused by Jacob of UKWorkshop

Jacob's sharpening technique, as discussed on the UKWorkshop forums, emphasizes traditional methods and freehand sharpening over the use of jigs and modern gadgets. He advocates for the use of oilstones and a minimalist approach to equipment, arguing that this method is both cost-effective and efficient when mastered.

Jacob's approach involves the use of a few oilstones to sharpen tools by hand, a technique he believes offers better control and results than using jigs. He often mentions that the over-reliance on gadgets and jigs, which became popular in the DIY boom of the 1980s, has overshadowed these traditional methods. His technique is rooted in consistent practice, developing muscle memory and skill, which he argues is more effective in the long run than using guides or jigs [oai_citation:1,sharpening technique | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening-technique.74627/) [oai_citation:2,sharpening technique | Page 6 | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening-technique.74627/page-6) [oai_citation:3,Sharpening | Page 5 | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening.144505/page-5).

He also stresses the importance of maintaining a flat surface on the sharpening stones, a critical aspect of ensuring a sharp and consistent edge on tools. Jacob often points out that this traditional method, while requiring more initial effort and practice, leads to a deeper understanding of the sharpening process and ultimately more precise results [oai_citation:4,Sharpening | Page 17 | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening.137789/page-17) [oai_citation:5,Sharpening | Page 10 | UK Workshop](https://www.ukworkshop.co.uk/threads/sharpening.137789/page-10).

For those interested in exploring his methods, it's advisable to start with a set of good quality oilstones and to practice regularly, focusing on achieving a consistent bevel angle and a keen edge through freehand techniques.
Fascinating! Flattering, but a very inaccurate representation of my views.
Actually quite worrying that it would be spot on with some details but also confidently misquote and misunderstand, so no change on previous comments on similar lines.
If AI was intelligent it would have asked me first!
 
Fascinating! Flattering, but a very inaccurate representation of my views.
Actually quite worrying that it would so confidently misquote and misunderstand, so no change on previous comments on similar lines.
If AI was intelligent it would have asked me first!
I was very surprised to get such a focused answer from the very superficial question I asked. Accurate or not to your actual views it does give a very good example of how AI is very good at mining data to extract information. This kind of use, well not sharpening methods, is what I use ChatGPT for the most. I treat it as a far better version of Google search where it gives me the information from multiple sources in one go. You can keep adding questions to refine the results.

As an example. Following the first answer I typed.

Are there any high profile woodworkers how disagree with Jacob’s approach (note the how instead of who -dyslexia- but it knew what I meant)

This was ChatGPTs answer

Several high-profile woodworkers advocate for sharpening techniques that differ significantly from Jacob's traditional, minimalist approach. These woodworkers often emphasize the use of modern tools, jigs, and a variety of sharpening systems to achieve precise and repeatable results.

**Michael Pekovich**, from Fine Woodworking, is a proponent of using a combination of waterstones and honing guides to sharpen plane blades and chisels. He stresses the importance of maintaining a flat back and using a side-clamp honing guide to ensure a consistent bevel angle. Pekovich argues that these tools make the sharpening process more reliable and accessible, especially for those who may not have developed freehand sharpening skills [oai_citation:1,Sharpening woodworking tools: 9 great articles - FineWoodworking](https://www.finewoodworking.com/2020/03/30/sharpening-woodworking-tools-nine-great-articles) [oai_citation:2,Woodworking 101: Basic Sharpening | Popular Woodworking](https://www.popularwoodworking.com/techniques/woodworking-101-basic-sharpening/).

**Bob Van Dyke**, also from Fine Woodworking, provides an overview of various sharpening systems, including oilstones, waterstones, and diamond stones. Van Dyke appreciates the consistency and efficiency that modern jigs and guides offer, which can be particularly beneficial for beginners or those who prefer a more systematic approach to sharpening [oai_citation:3,Sharpening systems explained - FineWoodworking](https://www.finewoodworking.com/project-guides/hand-tools/sharpening-systems-explained).

**The editors of Popular Woodworking** often recommend using grinding jigs and buffers for sharpening, especially for more complex tools like gouges and skews. They emphasize that while freehand sharpening can be effective, jigs provide a way to achieve precise and repeatable angles, which is crucial for maintaining the performance of tools over time [oai_citation:4,AW Extra 11/28/13 - Bowl Gouge Sharpening | Popular Woodworking](https://www.popularwoodworking.com/projects/aw-extra-112813-bowl-gouge-sharpening/) [oai_citation:5,Sharpening with a Grinding Jig and Buffers - Woodworking | Blog | Videos | Plans | How To](https://www.woodworkersjournal.com/sharpening-grinding-jig-buffers/).

These perspectives highlight a broader acceptance of modern sharpening aids and techniques within the woodworking community, contrasting with Jacob's traditionalist stance that relies heavily on skill development through practice and minimal equipment.
 
....

These perspectives highlight a broader acceptance of modern sharpening aids and techniques within the woodworking community, contrasting with Jacob's traditionalist stance that relies heavily on skill development through practice and minimal equipment.
This last para is totally wrong. Trad sharpening is easier and requires less skill development - but more information on how to do it, rather than just picking it up and finding out (or not) over time, which is how most of us had to learn.
Actually the jumble of information is pretty much what you'd get if you googled enough in the ordinary way. What is misleading is the air of confidence, as though it is an intelligent researcher. As others have said, AI is not intelligent, but it can fake it.
 
This last para is totally wrong. Trad sharpening is easier and requires less skill development - but more information on how to do it, rather than just picking it up and finding out (or not) over time, which is how most of us had to learn.
Actually the jumble of information is pretty much what you'd get if you googled enough in the ordinary way. What is misleading is the air of confidence, as though it is an intelligent researcher. As others have said, AI is not intelligent, but it can fake it.
Jacob. I’m very happy for you. There is another ‘person’ in the world you can disagree with 😜
 
For me the term artificial intelligence is misleading just as much as calling a phone smart or thinking there is anything social about social media.

In what sense are they defining intelligence ? I would suggest that it has to be a lot more than pattern matching or just logical comparisons and that for a minimum you would just state the what and then the intelligence should solve the how.
It would be interesting to hear how you define intelligence.
Professor Stuart Russell, who presented a radio series on the subject of AI, was asked if he thought it would ever surpass human intelligence. His response (and I paraphrase..) was to ask the questioner if they really believe that the arrangement of atoms in the human brain was the optimum arrangement that could ever be achieved.
 

Latest posts

Back
Top