AI and Engineering Design

[RichardM]

That may be why I spend most of my riding time on 1950's AJS and Matchless bikes: thick comfy seat, Jampot shock absorbers and teledraulic forks - what more would I want
View attachment 523605

So the Triumph is relegated to a poor 4th. Taking into account the issues with carbs also, maybe time to get AJS or Matchless riding suit?

 

[dave h 2]

What you need is a thermoregulation layer and a lightweight carbon composite base.

Looking at the original designs again in the fresh light of day, it looks like it’s just attached some meaningless buzzwords to various parts of a diagram of a seat, rather than actually designed anything. It's like what a management team would come up with if you asked them what a seat should be.

Yeah, but no, the management team would start with the price point, some of them would have no idea what a seat would or should be,

 

[Kayos]

Why does everyone use ChatGpt? Gemini and Claude are better for many things. I use all of them constantly.

I can't get on with Gemini, I do like Claude though

I have used Chatgpt since it was new, and it needs a certain way of prompting to get the best results

I will ask Gemini/Claude for a second opinion in one of my use cases

 

[fizzy]

I can't get on with Gemini, I do like Claude though

I have used Chatgpt since it was new, and it needs a certain way of prompting to get the best results

I will ask Gemini/Claude for a second opinion in one of my use cases

Gemini wasn't very good in the early days. 2.5 was much better but Gemini 3 is awesome. As is Claude 4.5 Opus.

 

[Farside]

I wonder if AI will ever be able to design a vehicle that can be easily worked on ?

I'd nominate the Hillman Avenger and Ford Escort Mk1 and 2 as the most easily worked-on cars of the 70s.
If they were to be brought back with modern tech, they'd be a hoot.

 

[Farside]

Breathable fabric? On a motorbike seat? What’s being “breathed out” from underneath?

Well, quite. The breathing is taken care of by airflow up top.

 

[Kayos]

Gemini wasn't very good in the early days. 2.5 was much better but Gemini 3 is awesome. As is Claude 4.5 Opus.

Gemini was the first (I think) you could give a document rather than copy/paste with limits. This was useful for reviewing and summarising

 

[Rannsachair]

I'd nominate the Hillman Avenger and Ford Escort Mk1 and 2 as the most easily worked-on cars of the 70s.
If they were to be brought back with modern tech, they'd be a hoot.

I would agree though I preferred working on a Saab 96 or Volvo Amazon.

 

[Dragnet]

Whilst I dont disagree with anything that @Screwdriver and everyone else has written above, the rhetoric around job losses sounds very familiar to the luddite movement of the industrial revolution.

As with any new technology, as it matures we will undoubtedly see a shift in the job market, but that has been going on since the start of time.

I can garuntee that all the jobs listed above have already been decimated by the use of computers and AI will just be the next step in that change.

50 years ago we had drawing offices full of people hand copying drawings, accountants with rows of people filling in ledgers by hand and couriers on bikes whipping between city offices with bits of paper. All of which are now long gone, replaced by CAD, Excel and E mail, and AI will just be the next step in that process.

Whilst I can't see into the future, I suspect that the limitations of AI will soon become apparent (its a tool not a solution) and there will always be a need for people to get their hands dirty, it may just look different to how it does today?

This is the assumption of the technological imperative, which the Luddites recognised even if it wasn't expressed in those terms. While accurate in one sense, what it misses is that there is more to the equation than what the tech-bros/C19th Titans of Industry say there is. It's hugely beneficial for the Musks and Salts of this world, but it removes human agency, delight, freedom and other rather more indefinable things than Profit. What is the point of this? The Luddites wanted to maintain a cottage industry that enabled them to have control of their labour, product and quality. The new mills created the slums that so affected Engels, and enabled Manchester to provide dhotis to the entire Indian subcontinent for a hundred years. I wonder what the Indian weavers thought of that. Are we increasing the capacity for human happiness?

Today we endeavour to seek out 'authenticity' by looking for small-scale production. It is heavily ironic that the 'artisanal sourdough' sold in upmarket bakeries is what would be considered normal bread in a rural household a couple of hundred years ago. Any creative person will have a preferred type of pen, pencil, paper, spanner, or computer programme because for humans it is about building connection. Richard Sennett's 'Homo Faber' is fundamental. We exist to make, and make ourselves in the process. What machine-learning is doing differently is giving a convincing impression of agency, which encourages people to treat it as though it has such agency. Instead of considering it as, say, a spreadsheet where checking the cells and calculations is fundamental to success, the output is treated reverentially. It is not being used as a tool, it is being used as an oracle.

As an architect, I know that a pencil drawing is a much better discussion tool with clients than a CAD printout. In the workshop, being able to divide fractional inches mentally means I can complete woodwork tasks in my head that friends who can only do metric have to get their phones out to calculate. We are making ourselves stupid by treating technologies as ends in themselves, rather than as means to remove drudgery - to perform repetitive, mundane or dangerous tasks for us. However, the way 'AI' is being discussed encourages the human to remove oversight totally. The output is not checked. The input is not checked. The reason for using it is not checked. The ecolopgical impacts are not checked. The emotional impacts are not checked. The societal impacts are not checked. It's a looming disaster.

An amusing adjunct is the Firefox addon 'Slop Evader'.

 

[Erie Fred]

Replacing people that really should not be in the engineering/design field to begin with.
AI has no ingenuity nor creativity, some engineers as well don't have either quality.


However, Not everything can be automated & dumbed down to a commodity.

Nor should be.

 

[jmp49]

I remember being at a afterparty at a hacking conference and there were people talking about AI and LLM's before it even became a thing in public discussions, and as hackers we all came to the same conclusion.
That training it wrong and injecting gibberish or bypassing things was going to open up a whole new class of attacks and methods to trick people and compromise things. And if you allow it to self train, it will train off other llm's output be it accidentally or deliberately and descend itself into slop and even more nonsense over time. And we're already seeing that in action.
That brave new world is already here.

https://www.anthropic.com/research/small-samples-poison

GitHub - Shiva108/ai-llm-red-team-handbook: AI / LLM Red Team Field Manual & Consultant’s Handbook

AI / LLM Red Team Field Manual & Consultant’s Handbook - Shiva108/ai-llm-red-team-handbook

github.com

I've got a lot more interesting reading matter, but you'll have to do your own research, this is part of how I make a living. Maybe you could ask chatgpt or claude to help.

 

[skotl]

Why does everyone use ChatGpt? Gemini and Claude are better for many things. I use all of them constantly.

I use ChatGPT for general queries ($20 a month plan) and Claude for coding (£200 a month)

 

[masleeve]

We are encouraged to use Copilot at work , it is a useful tool to help refine code etc, it does make mistakes and can disappear down a rabbit hole . Where it does excel is management speak both in decoding and writing - it will be getting used for objective setting .
Personally I have used Chat GPT , Gemini and Claude and I see them as useful tools