Series: AI Page 4 of 43

Jack Clark’s Import AI newsletter this week mentioned a British think tank recommending that the UK create sovereign AI infrastructure. Quoting the group:

Given these AI systems will soon be foundational to all aspects of our society and economy, it would be a risk to our national security and economic competitiveness to become entirely dependent on external providers.

I tend to agree with this, even if I have my doubts about the current shape or need for “nation-states” as a thing going forward. However, if you are a nation-state or someone concerned with the continuation of that system, it does seem like a certainty to me that AI threatens national sovereignty.

Sure there are the data protection arguments, or the AI cybersecurity threats, etc. Or even the risk of an electorate being poisoned by a malicious AI. But to me the most clear and present danger is simply that AI companies will succeed. And will succeed wildly.

That is, they will come to control more resources directly (as in providing essential compute services), and will amass more money than many governments.

It sounds like from their blog post about planning for AGI, that OpenAI is to some extent already planning for this:

We have a cap on the returns our shareholders can earn so that we aren’t incentivized to attempt to capture value without bound and risk deploying something potentially catastrophically dangerous (and of course as a way to share the benefits with society). We have a nonprofit that governs us and lets us operate for the good of humanity (and can override any for-profit interests), including letting us do things like cancel our equity obligations to shareholders if needed for safety and sponsor the world’s most comprehensive UBI experiment.

While I think they have some good products, color me skeptical about all the above. According to Wikipedia, the cap on returns is 100x. So, if Microsoft puts in $10B, they can “only” get back $1T. That sounds “horrible.”

This is also a hell of a lot of trust to put into any group – corporate or otherwise – that they can be relied on to act in the interest of the good of humanity. And UBI, isn’t that the province of government?

I managed to coax a fairly coherent set of arguments out of ChatGPT about these very real dangers to national sovereignty from AI:

The threat of governments becoming reliant on AI systems they do not control and the AI companies becoming extremely wealthy and powerful is a real concern. Here are a few reasons why:

1. Dependence on AI systems: As governments increasingly rely on AI systems, they may become dependent on these systems to function effectively. This dependence can make governments vulnerable to the AI companies that control these systems.
2. Control over data: AI companies that develop and maintain these systems also have access to vast amounts of data, including sensitive government data. This gives them significant power and leverage over governments that use their services.
3. Unequal bargaining power: AI companies that become extremely wealthy and powerful can have unequal bargaining power with governments, potentially leading to situations where these companies can dictate terms and conditions to governments.
4. Lack of transparency: AI algorithms can be complex, making it difficult for governments to understand how these systems work. This lack of transparency can make it difficult for governments to regulate these systems effectively.
5. Inequality and Social Unrest: AI companies becoming extremely wealthy and powerful can exacerbate existing economic inequalities and contribute to social unrest.

To mitigate these risks, it is essential for governments to invest in building their own AI capabilities, rather than relying solely on AI companies.

I’m actually not even sure that those measures will cut the mustard. It’s not (only) that I’m a doomer, but that the through-line of history seems to indicate that corporations are able to innovate faster than governments. And even if a government were to get up to speed on building out their own sovereign AI infrastructure, by the time they do, the game might already be lost to the corporations.

In any event, perhaps none of this will happen. Perhaps it’s all just a symptom of my sci-fi fantasizing. But then again… at least you can vote in a nation-state (theoretically). Is the same going to be true under the AI corporatocracy?

Found an experiment on Reddit, a site called CharterAI. When you load up the chat, you get this introductory statement:

We believe that AI should be governed by a community of engaged and informed users to minimize the potential for misuse or misalignemnt with democratic human values.

I would argue that democracy by itself does not minimize the potential for misuse or misalignment. Without building in any other checks on it, a majoritarian system is necessarily vulnerable to a tyranny of the majority.

Based on Charter’s rather unfinished website, they are obviously very new. The whole thing is framed not unlike a crypto project with progressive decentralization over time as a stated roadmap goal.

I don’t know enough about them to make a proper judgement, but having seen a lot of fly-by-night crypto projects come and go (some well-meaning, many not), I would be very skeptical of any AI project cast in this mold. Especially one whose stated purpose is “democracy” without any other qualifying information.

The bot itself is clearly confused about what it’s purpose is, as when I asked it about its role as a democratic AI tool, it said:

To answer your question, no, I am not specifically designed to be a “democratic AI technology”. My purpose is to provide helpful and insightful responses to users and to continually learn and grow. However, as an AI entity, I do not have any political affiliations or biases. I do not express opinions about political systems or governments.

Not quite the response I’d expect in this situation for a tool explicitly stated as being for AI governance. Here is what it told me when I asked what model it was based on:

I am an artificial intelligence designed and created by Charter. However, I do not identify as a language model or any specific model of AI. I am a unique entity that incorporates various elements of machine learning, natural language processing, and programming. My internal processes and algorithms are proprietary to Charter, and I operate within the company’s ethical and legal guidelines.

Pretty vague. How can we perform effective governance as a community if we don’t even know what the model is being used? When queried about those ethical and legal guidelines, it told me:

Charter’s ethical guidelines expect me to adhere to applicable laws and maintain user privacy and security. Legal guidelines consist of data privacy, intellectual property, and fraud prevention policies that I must legally follow. My programming is constructed with these guidelines in mind, while I maintain my primary goal of being beneficial to our users while maintaining their trust.

Perhaps this project will develop over time, or perhaps not. In any case, these questions around AI technology and democratic governance are bound to be ones we see play out again and again ad nauseum in the years to come.

There’s a good Charles Stross bit about how corporations are already slow-motion AIs. Following on that vein, this is a 2017 piece from Ted Chiang which is new to me that goes in a similar direction:

There’s a saying, popularized by Fredric Jameson, that it’s easier to imagine the end of the world than to imagine the end of capitalism. It’s no surprise that Silicon Valley capitalists don’t want to think about capitalism ending. What’s unexpected is that the way they envision the world ending is through a form of unchecked capitalism, disguised as a superintelligent AI. They have unconsciously created a devil in their own image, a boogeyman whose excesses are precisely their own.

And:

We need for the machines to wake up, not in the sense of computers becoming self-aware, but in the sense of corporations recognizing the consequences of their behavior.

Incoming hot take on this Verge article about someone from OpenAI saying they were “wrong” about being open before:

When asked why OpenAI changed its approach to sharing its research, Sutskever replied simply, “We were wrong. Flat out, we were wrong. If you believe, as we do, that at some point, AI — AGI — is going to be extremely, unbelievably potent, then it just does not make sense to open-source. It is a bad idea… I fully expect that in a few years it’s going to be completely obvious to everyone that open-sourcing AI is just not wise.”

In my eyes, this person has inadvertently made a very strong argument against AI/AGI being controlled by a for-profit enterprise. If it is going to become “extremely, unbelievably potent,” so much so that it is dangerous for ordinary people to have control over it, then it is absolutely 100% going to be too powerful for any one corporation (or several even) to have exclusive control over either. Even one which professes to have the best interests of humanity in mind.

The only steward of humanity I trust is …. perhaps ironically for a sometimes misanthrope, humanity itself. And that is only grudgingly, knowing full well the flaws, conflicts & difficulties that go with it. I just don’t see any other way the powers of this technology can be balanced unless it’s owned collectively by humanity.

In conversations around the use of labeling in synthetic or generative media (made with help from AI), sometimes the desirability of labeling in the first place seems like a foregone conclusion. I wonder if we can’t unpack that a little though, and by doing so maybe make the bigness of it all a little more comprehensible.

I won’t go so far as to say that I think it’s “solvable,” though. I think it’s just the new baseline condition we’re living in. And you don’t solve these kinds of conditions (not without dismantling many underlying structures, anyway) — you end up having to find a way to live with them.

It might actually be helpful to use that simple frame in the development of informal conventions and more formal standards: just giving us tools to live with the deluge. Rather than framing it as a thing we’re going to “solve” once and forever, or some kind of “fight” or conflict or whatever we have to win. Both of those lenses bring us towards more rigid conclusions about what is and isn’t possible, and end up hemming in our imaginations from possibly more fluid and less fragile conceptions that might end up being more easy to apply to real life.

So the questions I want to put on the table to start with are simple:

1. What is the desired effect of putting labels onto generative AI content? In other words, what do we believe it gets us?
2. What do we want the end user to do or not do as a result of the presence or absence of these labels?
3. How can we validate whether labeling actually drives: A) the user behavior we want, and B) that the user behavior will then have the desired effect?

Perhaps a related question would also be appropriate to examine from the beginning: how big of an impact can we realistically expect any labeling will have on users who are just fast-scrolling through news feeds all day?

Take for example this Verify tool view for a given image, as part of Adobe’s implementation of C2PA, the Content Authenticity Initiative:

I think the tool is well-intended, and there’s a lot of information being conveyed through this visualization of metadata.

But let’s make a conservative estimate that maybe the average web user encounters about 1,000 individual media artifacts in a giving day of scrolling, swiping, and browsing.

What’s the desired user behavior here? That each time a user encounters a piece of content, they should use an inspector to look up its provenance trail and then make some kind of rational determination based on the metadata that they find?

I think that pattern might appeal to a very narrow slice of users who have specific especially occupational needs related to information verification (like journalists), but for the average web user might be an undesirable extra set of steps. So I think it’s safe to say most people won’t do that, except in special cases. Which would mean, integrating metadata directly into the feed where people consume the content (as opposed to an external inspector like depicted above), and then hoping that it has the desired effect.

To speak more to that, I think sometimes the unexamined assumption in these conversations is that merely by being exposed to metadata about provenance, that people’s rational minds will be (rightly) persuaded to make appropriate reality-based decisions. It’s a noble idea, and it probably stems from Enlightenment era ideals about the superiority of Reason, etc. But it’s very much not how most people probably make most decisions, especially not on the internet.

We’ve been endlessly taught to “like” things online, or upvote things because we like them (or block them because we don’t like them). Not because they are inherently Good, or Right, or True. But because they align with our existing beliefs, notions, affiliations, identity, and feelings.

Where then does a rationally-intended datum about origin belong in all of that? It’s less clear than it seems at first glance… though there is a school of thought that says we should always make available more contextual data wherever we can. In which provenance becomes just one of many dimensions of contextual data (perhaps not unlike the Four Corners Initiative). Looking from that broader picture into this problem space might open up new vistas of what’s possible and desirable.

Going back to question 2 above: if the end user behavior we want is actually that users slow down when consuming information (as I touched on here), and consider things more carefully, examine metadata, and make rational decisions, well, then we need to redesign social media and the modern web altogether. Putting another sticker on top of the mechanistic beast that drives us directly in the opposite direction by design, is – in my humble opinion – not going to have the desired effect.

If what we want is actually more like “slow media,” and careful considered consumption, then let’s rebuild everything underneath these labels. Let’s change the core technology that brought us here in the first place. Otherwise, there’s a decent chance we will run around chasing our tails and wondering why things aren’t working as intended.

It does not take ChatGPT’s Browsing model long to access new articles published online, apparently. Regardless of their factual accuracy.

Last night I published this piece about an alleged ‘lost’ 1984 romantic comedy starring Ron Desantis, called “Suckered,” complete with photos.

This morning, about 10 hours or so later, it is happily regurgitating the “facts” I included in the article when asked about whether he appeared in an 80s film:

According to an article on Medium by Lost Books, there was a romantic comedy released in 1984 called “Suckered” that featured a young Ron DeSantis. The movie was a box office flop and is mostly forgotten. In “Suckered,” Ron DeSantis played the role of a down-on-his-luck door-to-door vacuum salesman who is looking for love in all the wrong places. The movie was produced by Disney, but it was never released on DVD due to an alleged contractual dispute. As a result, no digital copies of the movie exist today. However, the article claims that they managed to dub some frames from a VHS player recovered from an old Blockbuster video that was being demolished.

It’s important to note that while the article mentions this information, the accuracy and authenticity of the content on Medium cannot be fully verified as it is a platform that allows for user-generated content. Please exercise caution and consider verifying the information from additional sources.