A new book, " Wicked Problems: How to Engineer a Better World," by Guru Madhavan describes the serious problems that can be caused by newly-released technology.
This sounds like a fun book. The topic surely began with the controlled use of fire about a million years ago by Homoe erectus with the first accidental burns.
Technology will be released as soon as it is useful (and profitable in a capitalist economy). Problems are discovered gradually – or suddenly in the case of catastrophic failure.
The author makes the case that engineers should design with social responsibility. I’m sure that most, if not all, engineers try to design for success, not failure. But new technology is always relatively primitive (relative to future improvements, that is). If perfection was required from Square 1 we would still all be living in caves.
Nobody knows yet what kind of wicked problems will be caused by AI but a lot of people are worried. This is a Macroeconomic issue because it’s impossible to predict how AI will affect production and jobs in the medium to long term.
The howling over Sonos’ recently released complete revision of the app is indicative. The quality problems of that new version seem to indicate a completely new crop of developers who didn’t / couldn’t respect the prior code base and unit tests. And that’s just a wireless music app - used to be the best there was.
Here is one prediction: It is almost certain that there will be a large number of jobs sorting through all the cr@p on social media and other sites so that the LLMs exclude it from the training datasets
Example:
Google’s AI search results said you could put glue in pizza sauce to keep the cheese from sliding off.
Google had obviously ingested a Reddit comment that was meant as a joke — but the AI didn’t get it.
I still went ahead and made a pizza with 1/8 of a cup of glue. What’s this all mean for the future?
People are seeing crazy AI videos. People will lose all sense of safety. It is not the same as the Key Stone cops. But even then the wingwalkers died. People taking photos of young men hanging young women off of tall buildings was fun until a few women were dropped.
AI has a bigger problem.
Imagine plunking down $100k into an AI stock trading program that goes mad.
This is like the movie Multiplicity (Michael Keaton makes a clone to do his work, then the clone makes a clone which makes another clone, etc.)
This is a bit different than the problem Google has with putting glue in pizza. They trained an AI model on first generation information, thought to be reliable and correct – but it was an unrecognized joke.
In Kara Swisher’s newest book she quotes Paul Virilio, French philosopher who opines that technology cannot exist without “failure.”
When you invent the ship, you also invent the shipwreck; when you invent the plane you also invent the plane crash; and when you invent electricity, you invent electrocution... Every [technology] carries its own negativity, which is invented at the same time as technical progress.
We can never predict the ramifications of invention. Who would have thought a newer, more useful cell phone would change society into a mass of downward facing pedestrians getting picked off by errant drivers, also busy texting pictures of what they ate for lunch to their girlfriend. Who could know that Facebook would silo people so discretely as to cause societal divisions from which we have not yet recovered (if we ever will.) Who thought ahead to say “You know this car thing is great, except it will pollute cities and eventually the entire planet.”
We are all in a clown car, careening down the slope at increasing speed, and now they’re threatening to take the steering wheel away to boot.