I thought ai would only affect white-collar jobs, and i didn't think even a craftsman like a printer could get hit。
It is also easy to repair printers, and the maintenance manual is sent to ai, which, in terms of troubleshooting, directly hangs those who have worked for decades。
The maintenance manual is more detailed than most people's experience and suffers from malfunctions, which directly allow ai to combine the maintenance manual analysis with a very high accuracy rate。
For example, in the case of the following malfunction, where there is a maintenance manual, ai is very efficient in judging the failure, and even if there is no basis for it, the content generated by ai is fixed。
The difficulty in repairing printers was never the replacement of accessories, but the failure check. A lot of people think that if the printer is broken, they can change the parts, but it's not that simple。
The printer's malfunction is strange, the paper may be stuck on the output wheel, worn on the output wheel, the sensor may be misreported and circuit boards may occasionally die..
Different brands, and different years of machine principles, such as hp1020p and 1020w, are called 1020, but different years of production, different structural designs, completely unconventional parts, and cannot simply replace tests like desktop computers。
Hp1020p
Hp1020w
The real test is the experience of the maintenance staff, as well as the capacity for logic and reasoning。
Because you have to analyze:
Why is the paper stuck here
What's wrong with the selenium drums and the projector
Did the carbon powder in the powder box make the sensor alert
Machine or electronic or software
Scratch skills don't mean failure is accurate
Dismantling machines can be sophisticated, but it's also the strength of humans, and i don't even need to think about it now, because the same model creates muscle memories。
Many have repaired printers for more than a decade or even decades, and the detached manoeuvres are very smooth, but failure is not particularly accurate。
Fault determination requires a great deal of logical thinking and reasoning, and that's why older maintenance staff, even if they're very skilled in dismantling machines, are vulnerable to miscalculation.
On the one hand, many of them are reluctant to apply printer principles in a systematic way, such as the electrostatic imaging of laser printers, which was invented by carlson and is often not known to them。
On the other hand, the ability to reason logically is a human constraint, and the wealth of experience may not necessarily be “perhaps”。
The greatest advantage of ai is to make a logical judgement about the integration of knowledge
Ai has perfected the short board of human beings and has a strong logic capability to analyse the principles of source imaging of malfunctions. The database covers official documents and maintenance manuals。
The graphics are powerful, and the process of checking the steps to see the drawings is faster than humans。
Of course, there's a premise here: ai relies on accurate input. You must provide a sufficiently accurate description of the failure, accompanied by a maintenance manual, PDF, or a cut-off chart, otherwise ai will not be able to make an accurate judgement。
Examples are:
Ask ai correctly: fuscler sc2022, hint 042-326, what's wrong with the maintenance manual
Ai-instant generation program: tell you that 042-326 is a powder channel block and give you detailed steps to clean up the camera and the powder tunnel。
Error asking ai: what about printer not printing
Ai can only generate a bunch of crap, almost useless。
Real case: fuji xerox sc2022 fault 042-326
Maintenance staff: the whole process may take hours to half a day to remove the camera and clear the powder channels to check the drums, sort out the carbon powder blocks and even replace the visual agent。
Ai support: provides fault code, machine type and manual screenshots, which in one minute can analyse the source of the problem and give a detailed sequence of operational steps and attention。
As can be seen, the muscle memory of the dismantling machine still requires human practice, but logical judgement and failure screening are at the core and are the strength of ai. With decades of experience, ai can kill in seconds。
With ai, white can become a teacher, take a picture of the failure, and ai can create a solution。
Of course, today ai is mainly ai without hands. If i had a hand, ai would have just judged the breakdown machine maintenance, and i might have lost my job..
So let's not just think about the impact on white-collar salaries, even the “handicraft work” of printer maintenance is already quietly rewritten by ai。
In the future, whoever can upgrade himself with the ai tool in hand and logic, who can hold on to his job and who can feel through his traditional experience, will eventually be eliminated。
