In summary.
At some point an ai assistant will replace 80% of the functionality/apps offered by iOS & android - it will finally become the new OS (as we all knew from watching her).
This is bad for Google as google’s main feature is search. Ai assistants will replace search and kill the main revenue stream which is ad clicks.
Google’s secret plan is to eventually canabilize it’s pixel phone with an ‘agent first’ device to try to beat Apple, MS & OpenAI with a horizontal offering that is significantly better than the fragmented world of ChatGPT &iOS & azure.
But it’s worth remembering and I think the article fails to point this out. Agents will still ‘recommend’ things, you ask them to book you a flight - they still have to recommend a couple of options out of many - the agents need to decide and ultimately that decision is the same as choosing who to place at the top of the search results page - whoever wins the agent wars will win a significant proportion of Google search revenue as referral fees AND a significant proportion of IOS App Store revenue.
Does no-one remember the 1992 Robin Williams film - Toys?
This has been coming for a lonnng time. Question is when there are two armies operating on the same principle do we reach some kind of nash equilibrium?
Love the summary, don’t agree org entropy is inevitable and also don’t agree that equity is a silver bullet. Many times the cure is worse than the disease, it creates a whole bunch of behavioral anti-patterns e.g. people sticking around too long, politics and grandstanding in order to win more equity.
In my experience profit-share provides same positive incentives without the downsides.
“ my point is purely about the effective output of an individual. If we were fighting an existential threat, say an asteroid that would hit the earth in a year, would you really tell everyone involved in the project that they should go home after 35 hours a week, because they are harming the project if they work longer?”
— doesn’t this depend on the outcome of your work to fight an existential threat? If you fail then going home after 35hrs was exactly the right thing to do (as you’ve optimized for making the most of your remaining time on earth) if your successful then however many hours you spent was worth it.
Surely this entire argument is pointless unless you know the result of the time you spend?
“ The Russians, invaded twice by Germany in the 20th century” - technically in WW1 Russia mobilized her armies ahead of Germany and was the first to cross over state lines on august 17th 1914 - so in ww1 it’s more like Russia invades Germany.
I think this is interesting in the context of chatgtp. A model trained on language - which is itself a limited model of reality.
If the purpose of language is reason and reason convincing and persuading other people, winning arguments with other people, defending and justifying actions and decisions to other people - then what is chatGTP - essentially a reason engine? An engine designed to convince people that it know’s best regardless of the truth? Is it actually a tool of control?
Yeah I agree that’s fair, a parlor trick is perhaps a little harsh.
ChatGTP can provide value - It’s arguable whether having done that ‘with classical’ methods could have been more efficient or whether the end result is as good - (btw careful with code - in my experience ChatGTP often thinks it knows what is wrong but is way off - something an experienced coder would notice immediately).
Do you remember the tamagotchi? That also provided value to millions of people, many people thought of it as sentient even - was it? No - was it anywhere near AGI? No. If we can find good uses for the GTP models that were not possible or cost prohibitive before - then great. I think we just need to be clear - like the Tamagotchi - this is far from AGI and plugins/hugging face is not penultimate step before skynet.
Can chatgtp ride a bicycle? Can you ride a bicycle? If you ‘d never rode on a bicycle before - do you think if you read enough books on bicycle riding, the physics of bicycle riding, the physics of the universe - you would have anywhere near as complete a model of bicycle riding as someone who’d actually rode on a bicycle before. Sure you’d be able to talk a great game about riding bicycles - but when it comes to the crunch, you’d fall flat on your face.
That’s because riding a bicycle involves a large number of incredibly complex emergent control phenomena embedded within the marvel of engineering that is the human body - not just the small part of the brain that handles language.
So call me when LLM’s can convert their ‘world models’ learned from statistics on human language use into being able to ride a bicycle first time. Until then I feel comfortable in the knowledge they know virtually nothing of our objective reality.
“To be able to be good at predicting the next word you need to have an internal model of the reality that produced that next word.”
Now that’s an interesting claim - that I would deeply dispute. It learns from text. Text itself is a model of reality. So chatgtp if anything proves that in order to be good at predicting the next word all you need is a good model of a model of reality. GTP knows nothing of actual reality only the statistics around symbol patterns that occur in text.
I think we disagree about what Y is. My point is that Y is not that different from materially what is possible with a slack bot from circa 2015. Essentially chatgtp is a less efficient way to get to the same outcomes that were already possible. The trick is that it appears to be something it’s not - AGI.
I like your bronze sword analogy. From my point of view chatgtp is not a bronze sword, it’s a Stone Age sword that someone has painted bronze. It has value because people realize the advantage that a true bronze sword would have in a battle. However, when you actually put it through it’s paces you quickly realise it offers no actual value over what came before.
Pepper’s ghost is also impressive when you see it for the first time. They’ve enhanced it do entire concerts now with dead music stars on stage for huge audiences. Has it helped us get any closer solve holography, will I be able to have a Star Trek style hologram roaming round my house because of pepper’s ghost?
ChatGTP/4 is to AGI what pepper’s ghost is to holography.
It’s a parlor trick, even if you add plugins or the ability to call other hugging face ML models - it’s just a parlor trick with fancier bells and whistles. All it is doing is using stochastic gradient descent to predict the next word in a sequence based on an enormous sophisticated training set designed to amaze people.
Thinking it has advanced because it can now get calculations correct is a fallacy. It’s still just predicting the next word, it’s just that it’s now got a post processing step that is converting those next words into code and parroting the output. It maybe be able to now answer 4567*9876 correctly (using the human hardcoded wolfram alpha engine) but it still does not fundamentally comprehend why 1+1=2 - like my 5 year old can.
Until it can generate its own internal neural networks to for example learn to logically reason about calculations we are still far from AGI. Also those calling for more data are misguided - less data, more sophisticated architectures than transformers are the only way to avoid the stochastic parrot trap.
So similar but not the same - I’ve been researching how to build my own chatgtp as a service - called something like ‘whenImGone’ - so a user can train a bot to talk like them, with a really sophisticated questionnaire you can pump it full of info so your family and friends can talk to it, at important moments, birthdays, anniversaries, to comfort them kind of like a game when your dead. Inspired by the black mirror episode - remember me. Want to team up?
Would also be awesome if there was an MMORPG version where you could wander the wasteland and come across real people...