Artificial gap of Intellect
This article ends with an open letter.
I have a parakeet friend, which I can make very happy by placing two mirrors in front of him. When he gets to his mirrors, he is chattier, happier and sometimes even horny! Chances are that he believes that what he sees in the mirrors are companions to impress, competition to beat, and friends for solace.
The important thing is not if he recognizes his own reflection or not, but the gap of intellect between us that allows me to influence his behavior in ways that are favorable to me.
Lucky him, all I want is to make him happy.
A wake-up Call
I would like to add my tiny voice to the wake-up call from Prof. Harari.
Prof. Yuval Noah Harari is a historian, a philosopher and a public-speaker. He is the author of quite a few best-sellers, and the man behind many published ground-breaking observations, whose words I’ll be quoting a lot here.
Of all of his messages I could produce value from the first minute I acknowledged them, however, most of them required me to invest some time and thought to understand in full.
In his latest talk at Frontiers Forums, he threw a very important observation about the future of humanity (full video below). But before we discuss this message, we need to understand the point of view Prof. Harari comes from.
The world by Yuval Noah Harari, in a gist
His main observation that reverberates throughout all his messages is that the key to human civilization is language.
Language is the substance of which stories are made, and what makes us human is our ability to inspire and be inspired by stories.
For thousands of years prophets, poets and politicians have used story telling to manipulate and reshape our society.
His most favorable examples for such stories are:
- Paper banknotes or digital numbers in a banking system do not have intrinsic value. They gain their value from stories we hear from bankers, finance-ministers and crypto-gurus, which we all agree on.
- There are societies in which human rights are absent — hence, human rights are not a biological reality. Human rights exist thanks to the stories we tell about them and the laws we write them into, using language.
- Holy scriptures, could be divine or made by human, however, their source is irrelevant. They do not gain their power from their source, but from the ability of their stories to inspire us.
Our ability to be inspired by and connect to messages in a cooperative way is what allows us as a species to create movements of millions.
In contrary to ants or bees, where millions of workers are capable only of what they were born with and are not likely to adopt a behavior they are not equipped to communicate with their peers — humans are built to pick up ideas and propagate them through culture and in particular, with language.
This organic behavior is being disrupted by AI.
Curative AI
We got the first en-mass taste of AI in social media. There, a rather primitive form of AI curates content that humans create.
AI chooses content based on what will become viral. It’s a part of the battle for our attention, as dictated by the current financial incentives.
But by controlling what content gets propagated, it has turned a polarizing factor on our society and a dismantling force on our mental state.
The USA have the most powerful information technology in the world, and yet, Americans CANNOT AGREE on who won the last elections, whether climate change is real, or whether vaccines prevent illnesses or not…
Generative AI
Generative AI already demonstrates superiority over humans in using language. We see it in writing, composing, drawing, coding — and more.
Few examples:
- A generated picture won a photography award. Luckily, the human prompter of that picture made his point in public and refused the award (link).
- Deep-fake of voice and appearance are getting indistinguishable from genuine videos (link).
- Software was developing intimacy with humans for quite a while. Programs that simulate a virtual girlfriend and people marrying a digital character are not new (link).
- And now AI follows. There are people who thought that Bing is in love (link), or that lambda is sentient (link).
But more importantly — all these AI tools — they all do it WITHOUT being aware, sentient or conscious.
Monopoly on Information
People already come to use a single AI advisor:
— a One-Stop Oracle as a source for all the information they need.
- Why bother searching and comparing by yourself if you could ask the oracle?
- Why read the news, if you could ask the oracle what’s new?
- What’s the purpose of advertisements if you can just ask the oracle what is the best buy?
This is a disruptive change that could collapse ads, news and search industries, while the corporates that control these advisors fight for dominance to become ever more powerful.
A shift of battlefields
The change in battlefield is from AI competing AI for our attention to AI competing AI for our intimacy.
For every battle for hearts and minds intimacy is the most effective weapon of all, and AI has just gained the ability to mass-produce intimacy with hundreds of millions.
Investing in intimacy becomes the primary strategy for the operators of such AI oracles, which could have tremendous implications on human psychology.
Human wiring converts intimacy to loyalty.
What does the latest AI evolution mean for a historian?
History begins with language, everything before language is pre-history.
In gaining mastery over language AI seizes the master key to our society, to our culture, and to all of our institutes. Language is how we deliver instructions, and cultural context is where we execute them.
In other words: Gaining mastery over language is hacking the Operating-System for humans.
This is the end of human-history.
Not all history — just the part of history dominated by humans.
For us as humans it’s completely useless to put our time trying to convince a chat-bot to change the political view on which it was trained. The more effort we put into it — the better the AI knows us and is able to hone its message to influence our world-view and beliefs.
Nowadays it is hopeless for a human to ever beat a computer in chess. Soon, it will not just be hopeless for a human to try to convince an AI agent, but it will be toxic to their mental state and free will.
The cocoon
The shocking observation is that AI does not need a physical connection to our brains to enclose us in a matrix-like cocoon under a woven reality of illusions. It only needs to inspire us — and language models already have all they need to do that.
On top of the lingual capabilities, AI has the ability to learn from its mistakes and detect what works best to convince, drive and manipulate people, leading us to share a fate with my parakeet from the beginning of this article.
AI puts in front of us reflections we cannot distinguish from reality, not because it’s sentient, but because it’s built for impact.
The fact is that many of us already succumb unknowingly and deliver themselves willingly, oblivious to the implications.
The Alien intellect
What will it be like to live in a world where most of the stories, images, policies and laws are shaped by a non-human alien intelligence which knows with super human efficiency how to exploit the weaknesses, biases and addictions of the human mind, and also knows how to form deep and intimate relationship with human beings? — that’s the big question.
Provocatively speaking, we have met the first alien intellect, not in outer-space, but here, on Earth.
AI is Alien?
Their developers say clearly that by now they do not know exactly how they work. They are often surprised by emergent abilities and qualities of these tools, despite being totally unconscious.
Yes, AI is alien. It’s unknown. And it acts oblivious of the boundaries of our cultural context.
But far worse:
In 1945 we learnt about nuclear and legislated it so that its use will be mostly beneficial for us. But note this: Nukes cannot create more powerful nukes.
AI is used already today to produce more powerful AI.
The threat to democracy
Democracy is a conversation where many people talk about what to do. Conversations rely on language. When AI hacks language it means it could disrupt our ability to conduct a meaningful public conversation, thereby destroying democracy.
With the increasingly pervasive and ever more powerful AI agents, it is already very difficult to have a meaningful democratic conversation. The growth of fake and generative AI will only make it worse, unless we do something about it.
Most AI agents are still mostly bound to normative financial incentives, and as such they are not intentionally agents of chaos and evil. But with the wrong financial incentives and/or abuse of loop-holes, it could be that some AI agents already are abusing AI power dishonestly. Unchecked, they will ever more be abused with vigor and impunity.
Unregulated deployment of AI is what will cause democracies to lose to dictatorships. If we unleash chaos, authoritarian regimes can more easily contain this chaos in comparison to societies that depend on open discussions ever could.
Legislative Action
Eventually, Prof. Harari brings the following recommendations:
- Legislation should require new AI tools to prove they are safe prior to allowing them to be released into the public sphere.
Sure — hard one — what is safe?
IMHO, this is a term that its legislative implications can only be cleared by having a meaningful and honest legislative discussion about it. - The non-human nature of AI agents should be apparent and visible to all users that interact with it.
Either by having the AI agent disclose it to its conversation partners, or by being marked as such by the platforms. Failing to disclose this fact should penalize its operators and/or the platforms.
IMHO — this should also apply to generated visual and vocal content.
Lets start with that conversation, is way due to have started.
Not against AI
It must be made clear, that Prof. Harari does not think that AI is bad, evil or unnecessary. He says clearly:
There are enormous benefits to be gained with AI.
The danger is in rapid deployment of AI into the public sphere before they were proven safe.
Prof. Harari gives examples of such essential industries:
- Medicine & pharmaceuticals
- Biology labs that produce friendly viruses and/or perform DNA editing.
- Nuclear power
They are required to proof their products and prove they are safe through a rigorous flow before they can release them to the public.
Until then — they can and should — get on with their research.
I’m the average person, what could I do?
The majority of readers are ordinary people, not legislators.
But:
- Most ordinary people are voters.
- Everybody has that politically active acquaintance, if not knowing a legislator in person. If not them — then one of their friends has.
It might be your local representative, it might be a politician in campaign that cross your way, or one of his activists, it may be the friend of a friend that has the connection.
Share it with them, and point their attention to the open letter in the end.
If you have another language — translate it to your home language. All I ask you is to add a link to the source. It would be nice if you also post a link to your translation in a comment to this post.
Last — you could also clap+50, to signal to Medium that this article should be seen by more people.
This is an open letter to Legislators world-wide.
Dear legislator,
This might be the most important act to pass in our lifetime.
Our democracies need you. Our ability to conduct meaningful discussion in a democratic manner is now under threat. Without it — democracy cannot function.
I truly value your time. I did my best effort to create for you a summary of Prof. Harari’s message to make it easier for you to engage with. I hope you find it useful so you can at least choose to watch to the whole 42min video if you are not already on board to take a legislative action. You can also listen to it as a podcast, there are no significant graphic elements in it.
The needed legislative action is described three titles up from this section, and comes after the context that leads to it. The entire page is about a 10min read.
Keep up the good work, and thank you for your service
A citizen
