A Future Worth Thinking About: Recent Episodes
None
Thinking about magic, cyborgs, robots, and artificial intelligence--and why some of those words could use changing--since 1982.
So, new research shows that a) LLM-type “AI” chatbots are extremely persuasive and able to get voters to shift their positions, and that b) the more effective they are at that, the less they hew to factual reality.
Which: Yeah. A bunch of us told you this.
Again: the Purpose of LLM- type “AI” is not to tell you the truth or to lie to you, but to provide you with an answer-shaped something you are statistically determined to be more likely to accept, irrespective of facts— this is the reason I call them “bullshit engines.” And it’s what makes them perfect for accelerating dis- and misinformation and persuasive propaganda; perfect for authoritarian and fascist aims of destabilizing trust in expertise. Now, the fear here isn’t necessarily that candidate A gets elected over candidate B (see commentary from the paper authors, here). The real problem is the loss of even the willingness to try to build shared consensus reality— i.e., the “AI” enabled epistemic crisis point we’ve been staring down for about a decade.
Other preliminary results show that overreliance on “generative AI” actively harms critical thinking skills, degrading not just trust in, but the ability to critically engage with, determine the value of, categorize, and intentionally sincerely consider new ways of organizing and understanding facts to produce knowledge. Further, users actively reject less sycophantic versions of “AI” and get increasingly hostile toward/less likely to help or be helped by other actual humans because said humans aren’t as immediately sycophantic. And thus, taken together, these factors create cycles of psychological (and emotional) dependence on tools that Actively Harm Critical Thinking And Human Interaction.
What better dirt in which for disinformation to grow?
The design, cultural deployment, embedded values, and structural affordances of “AI” has also been repeatedly demonstrated to harm both critical skills development and now also the structure and maintenance of the fabric of social relationships in terms of mutual trust and the desire and ability to learn from each other. That is, students are more suspicious of teachers who use “AI,” and teachers are still, increasingly, on edge about the idea that their students might be using “AI,” and so, in the inimitable words and delivery of Kurt Russell:
Combine all of the above with what I’ve repeatedly argued about the impact of “AI” on the spread of dis- and misinformation, consensus knowledge-making, authoritarianism, and the eugenicist, fascist, and generally bigoted tendencies embedded in all of it—and well… It all sounds pretty anti-pedagogical and anti-social to me.
And I really don’t think it’s asking too much to require that all of these demonstrated problems be seriously and meticulously addressed before anyone advocating for their implementation in educational and workplace settings is allowed to go through with it.
Like… That just seems sensible, no?
The current paradigm of “AI” encodes and recapitulates all of these things, but previous technosocial paradigms did too, and if these facts had been addressed back then, in the culture of technology specifically and our sociotechnical culture writ large, then it might not still be like that, today.
But it also doesn’t have to stay like this. It genuinely does not.
We can make these tools differently. We can train people earlier and more consistently to understand the current models of “AI,” reframing notions of “AI Literacy” away from “how to use it” and toward an understanding of how they function and what they actually can and cannot do. We can make it clear that what they produce is not truth, not facts, not even lies, but always bullshit, even when they seem to conform to factual reality. We can train people— students, yes, but also professionals, educators, and wider communities— to understand how bias confirmation and optimization work, how propaganda, marketing, and psychological manipulation work.
The more people learn about what these systems do, what they’re built from, how they’re trained, and the quite frankly alarming amount of water and energy it has taken and is projected to take to develop and maintain them, the more those same people resist the force and coercion that corporations and even universities and governments think pass for transparent, informed, meaningful consent.
Like… researchers are highlight that the current trajectory of “AI” energy and water use will not only undo several years of tech sector climate gains, but will also prevent corporations such as Google, Amazon, and Meta from meeting carbon-neutral and water-positive goals. And that’s without considering the infrastructural capture of those resources in the process of building said data centers, in the first place (the authors list this as being outside their scope); with that data, the picture is worse.
As many have noted, environmental impacts are among the major concerns of those who say that they are reticent to use or engage with all things “artificial intelligence”— even sparking public outcry across the country, with more people joining calls that any and all new “AI” training processes and data centers be built to run on existing and expanded renewables. We are increasingly finding the general public wants their neighbours and institutions to engage in meaningful consideration of how we might remediate or even prevent “AI’s” potential social, environmental, and individual intellectual harms.
But, also increasingly, we find that institutional pushes— including the conclusions of the Nature article on energy use trends— tend toward an “adoption and dominance at all costs” model of “AI,” which in turn seem to be founded on the circular reasoning that “we have to use ‘AI’ so that and because it will be useful.” Recurrent directives from the federal government like the threat to sue any state that regulates “AI,” the “AI Action Plan,” and the Executive Order on “Preventing Woke AI In The Federal Government” use term such as “woke” and “ideological bias” explicitly to mean “DEI,” “CRT,” “transgenderism,” and even the basic philosophical and sociological concept of intersectionality. Even the very idea of “Criticality” is increasingly conflated with mere “negativity,” rather than investigation, analysis, and understanding, and standards-setting bodies’ recommendations are shelved before they see the light of day.
All this even as what more and more people say they want and need are processes which depend on and develop nuanced criticality— which allow and help them to figure out how to question when, how, and perhaps most crucially whether we should make and use “AI” tools, at all. Educators, both as individuals and in various professional associations, seem to increasingly disapprove of the uncritical adoption of these same models and systems. And so far roughly 140 technology-related organizations have joined a call for a people- rather than business-centric model of AI development.
Nothing about this current paradigm of “AI” is either inevitable or necessary. We can push for increased rather than decreased local, state, and national regulatory scrutiny and standards, and prioritize the development of standards, frameworks, and recommendations designed to prevent and repair the harms of “generative AI.” Working together, we can develop new paradigms of “AI” systems which are inherently integrated with and founded on different principles, like meaningful consent, sustainability, and deep understandings of the bias and harm that can arise in “AI,” even down to the sourcing and framing of training data.
Again: Change can be made, here. When we engage as many people as possible, right at the point of their increasing resistance, in language and concepts which reflect their motivating values, we can gain ground towards new ways of building “AI” and other technologies.
There’s a new open-access book of collected essays called Reimagining AI for Environmental Justice and Creativity, and I happen to have an essay in it. The collection is made of contributions from participants in the October 2024 “Reimagining AI for Environmental Justice and Creativity” panels and workshops put on by Jess Reia, MC Forelle, and Yingchong Wang, and I’ve included my essay here, for you. That said, I highly recommend checking out the rest of the book, because all the contributions are fantastic.
This work was co-sponsored by: The Karsh Institute Digital Technology for Democracy Lab, The Environmental Institute, and The School of Data Science, all at UVA. The videos for both days of the “Reimagining AI for Environmental Justice and Creativity” talks are now available, and you can find them at the Karsh Institute website, and also below, before the text of my essay.
All in all, I think these these are some really great conversations on “AI” and environmental justice. They cover “AI”‘s extremely material practical aspects, the deeply philosophical aspects, and the necessary and fundamental connections between the two, and these are crucial discussions to be having, especially right now.
Hope you dig it.
Reimagining “AI’s” Environmental and Sociotechnical MaterialitiesDamien P. WilliamsUNC Charlotte
There are numerous assumptions bundled into the current thinking around what “artificial intelligence” does and is, and around whether we should even be using it and, if so, how. Those pushing “AI” adoption tend to presuppose it necessarily will be good for something— that it will be useful and solve some problem— without ever defining exactly what that problem might be. Often, we see that there are these pushes towards paradigms of efficiency and ease of work and “rote” tasks being taken off our hands without anyone ever asking the fundamental follow-up question of “…okay but does it actually do any of that?” Relatedly, it’s often assumed that “artificial intelligence” will become or will make other things “better” in some nebulous way if only we just keep pushing, just keep building, just keep moving towards the next model of it. If we keep doing that, then eventually, we’re assured, “in just ten years,” “AI” will turn into the version of itself that will solve all our problems. But this notion that in ten years, “AI” will be embedded in everything and will be inescapable and perfect is something we’ve been hearing for the past 50 years.
This recurrent technosocial paradigm of “AI Summer” and “AI Winter” exists for a reason; these hype-cycles pushing towards automation, neural nets, big data, or algorithms over and over again represent externalities which must be addressed in a deeper way through questions like, “What are the values of the people who push ‘AI’s’ ‘inevitability,’ and what are their actual goals?” Because, while people might think they mean the same things when they say “AI,” or are indicating the same kinds of needs to be met, in truth, we’re very often talking past each other. Without a clear understanding of what it is we each and all actually think of as the “good” of “AI” technology— without confronting that question in a very direct and intentional way— different groups will just keep pushing in different directions, and whoever has the predominant access to and control over the levers of power wins the right to define the problems that “AI” seeks to address. But in many cases, those are problems they and their vision of “AI” helped to create.
Current estimates hold that water consumption increased ~34% in areas where Microsoft and Google placed datacenters for search and “AI,” and that every email’s worth of text you have an LLM “AI” write consumes a pint of water. Put another way, imagine if every time you composed 150 of your own words, you had to just take out a 16 oz water bottle, fill it up, and dump it in the trash. We’re not just talking about water for cooling servers, either. In thermal power plants, you need water to turn into steam to run turbines, and then to cool the systems which do that, as well. So the more energy needed, the more water used in production and cooling. And while many highlight that some systems only use this water once and then release it, even that is a process and a period of capturing that water, both removing the water from use, and potentially trapping and killing organisms living in it. Additionally, the water returned after the “once through” process has a significantly higher temperature than when it started. It should be said that the numbers in this discussion are estimates based on known figures for chip performance, electricity production, and whatever data’s been wrenched from “AI” corporations. They’re estimated because these companies do not release their actual resource consumption numbers.
Further, the data centers that support “AI” are oftentimes built in communities that are already resource scarce, and pulling water from or putting emissions into these communities ensures that “AI’s” harms are necessarily disproportionately enacted on the people who can least afford to bear them. Rather than rulemakers just paying lip-service to people’s grievances, logging them in a repository somewhere, and making whatever rules they intended to make to begin with, both the creation and regulation of “AI” must be directed by those whom it’s most likely to harm. But while marginalized communities absolutely must have meaningful input when it comes to technologies which will be wielded against them, there also has to be a centralized response in the form of some standard-setting body. And, recursively, that standard-setting body will have to be meaningfully responsive to the needs of those most likely to be harmed if said regulations and standards go wrong.
And so, we have to ask our questions: Who is most harmed by current uses of “AI”? What does the energy footprint of a data center actually look like? How much water and fossil fuel does it take to run “AI’s” servers and their computations? What are their carbon and waste heat emissions? Because the more we dig down on this, the more we truly confront the next questions: Should we be doing “AI” differently? What would it take to build “AI” in a different way? What would it take to power “AI” in a truly renewable way? And what and whom do we even want “AI” to be for? If it helps, you can try to think of it as a game:
First major “AI” firm to use only renewable energy sources, an open source and radical consent model for the collection and use of training data, and a community partnership regulatory process which centers and heeds the needs of the most marginalized, wins.
Suggested Citation:
Williams, Damien P. “Reimagining ‘AI’s’ Environmental and Sociotechnical Materialities,” appearing in Reimagining AI for Environmental Justice and Creativity, Reia, J., Forelle, MC and Wang, Y. eds. Digital Technology for Democracy Lab, University of Virginia. 2025. https://doi.org/10.18130/03df-zn30.
It’s really disheartening and honestly kind of telling that in spite of everything, ChatGPT is actively marketing itself to students in the run-up to college finals season.
We’ve talked many (many) times before about the kinds of harm that can come from giving over too much epistemic and heuristic authority over to systems built by people who have repeatedly, doggedly proven that they will a) buy into their own hype and b) refuse to ever question their own biases and hubris. But additionally, there’s been at least two papers in the past few months alone, and more in the last two years (1, 2, 3), demonstrating that over-reliance on “AI” tools diminishes critical thinking capacity and prevents students from building the kinds of foundational skills which allow them to learn more complex concepts, adapt to novel situations, and grow into experts.
Screenshot of ChatGPT[.]com/students showing an introductory offer for college students during finals; captured 04/04/2025
That lack of expertise and capacity has a direct impact on people’s ability to discern facts, produce knowledge, and even participate in civic/public life. The diminishment of critical thinking skills makes people more susceptible to propaganda and other forms of dis- and misinformation— problems which, themselves, are already being exacerbated by the proliferation of “Generative AI” text and image systems and people not fulling understanding them for the bullshit engines they are.The abovementioned susceptibility allows authoritarian-minded individuals and groups to thus further degrade belief in shared knowledge and consensus reality and to erode trust in expertise, thus exacerbating and worsening the next turn on the cycle when it starts all over again.
All of this creates the very conditions by which authoritarians seek to cement their control: by undercutting the individual tools and social mechanisms which can empower the populace to understand and challenge the kinds of damage dictators, theocrats, fascists, and kleptocrats seek to do on the path to enriching themselves and consolidating power.
And here’s OpenAI flagrantly encouraging said over-reliance. The original post on linkedIn even has an image of someone prompting ChatGPT to guide them on “mastering [a] calc 101 syllabus in two weeks.” So that’s nice.
No wait; the other thing… Terrible. It’s terrible.
Screenshot of a linkedIn post from OpenAI’s chief marketing officer. Captured 04/04/2025
Understand this. Push back against it. Reject its wholesale uncritical adoption and proliferation. Demand a more critical and nuanced stance on “AI” from yourself, from your representatives at every level, and from every company seeking to shove this technology down our throats.
Back in October, I was the keynote speaker for the Society for Ethics Across the Curriculum‘s 25th annual conference. My talk was titled “On Truth, Values, Knowledge, and Democracy in the Age of Generative ‘AI,’” and it touched on a lot of things that I’ve been talking and writing about for a while (in fact, maybe the title is familiar?), but especially in the past couple of years. Covered deepfakes, misinformation, disinformation, the social construction of knowledge, artifacts, and consensus reality, and more. And I know it’s been a while since the talk, but it’s not like these things have gotten any less pertinent, these past months.
As a heads-up, I didn’t record the Q&A because I didn’t get the audience’s permission ahead of time, and considering how much of this is about consent, that’d be a little weird, yeah? Anyway, it was in the Q&A section where we got deep into the environmental concerns of water and power use, including ways to use those facts to get through to students who possibly don’t care about some of the other elements. There were a honestly a lot of really trenchant questions from this group, and I was extremely glad to meet and think with them. Really hoping to do so more in the future, too.
Me at the SEAC conference; photo taken by Jason Robert (see alt text for further detailed description).
Below, you’ll find the audio, the slides, and the lightly edited transcript (so please forgive any typos and grammatical weirdnesses). All things being equal, a goodly portion of the concepts in this should also be getting worked into a longer paper coming out in 2025.
Hope you dig it.
Until Next Time.
Audio:
http://afutureworththinkingabout.com/wp-content/uploads/2024/12/DPW_-_SEAC-Keynote_2024_10_04.mp3Slides:
DPW_-_TVKD-GenAITranscript:
0:03
I’m going to turn this mic on because I’m probably going to wander a little bit, and this will help make sure that everybody can still hear me, even as I do.
Thank you very much for that introduction. I really appreciate it. I really appreciate all of you for being here today. I appreciate you taking your time and joining me at lunch. This is going to be a talk that’s going to talk about the kind of sociotechnical values and impacts of “AI” generative, “AI” in particular, more than it talks about the specific technical details of “AI”. That being said, one of the things that we are going to talk about, just to get us started, is going to be a little bit, a very little bit, about how generative “AI” works when we talk about things like GPTs, large language models, or any number of other things that we put under the header of “AI” writ large, what we are talking about is in a very real sense of kind of a system of interoperable algorithms that are used to weight, arrange, sort and return large associations of language large language models are the kind of overarching framework under which things like GPTs to text based transformers for OpenAI, things like Google, Gemini, these Are the frameworks on which they work.
Large language models can also be applied to things like text generators. They can be applied to video generators, and in the process of doing that, what they do is they take that large language of large language batch of associated data, and they then frame it and associate it, tag it to video images or audio with which they have also been trained. And so the process is about creating a greater fidelity of statistical correlation between text and text, text and image, text and video or text and audio. In order to do any of this work, what you have to do is you have to have human beings go in and also engage in a process of what’s known as reinforcement training. That reinforcement process oftentimes requires a large group of humans to teach these systems whether they’ve gotten it right or whether they’ve gotten it wrong, and then to nudge them closer and closer towards getting it right. The question of, “What does right mean in this context and who is doing that training?” undergirds all of this. The data that large language models use can be in the billions or now even trillions, of tokens large. And when we say token, what we mean in that space is a word or a phrase.
The word “token” is used to talk about a word or discrete chunk of language which is associated in a particular way. At the top of this slide, what I have is what is a very simple vector map. It is a gift that shows the associations between certain words within a vector space. This is how the original Forerunner language models things like GLoVe or Word2Vec which were used up until about 2016 as the kind of cutting edge of language modeling in the “AI” space. This is how you could see them doing the work that they do. And if you take a look at the particular map that I’ve provided here, the one that I’ve chosen that I use for most of the conversations I have around this space, you’ll probably notice a very particular set of associations happening in this you see the word man and the word King are mapped in a vector space. That is, they are related to each other closely. When you plot out how words are connected to each other, what connotations words have, these show up very near each other, and the word woman and the word King show up very near each other in that vector space. The thing of it is, is that when you look at the vector space, the word “woman” doesn’t just appear next to the word “king”— or “queen,” rather, and vice versa.
The word “woman” appears next to the word “secretary,” very closely in the vector space, the word “nurse,” the word “teacher.” And it appears very distant from the words president, doctor, CEO. This is a visual representation of how biases get embedded within language modeling systems, the data on which they are trained comes from sources which have those biases encoded in them. This has become a kind of a standard understanding throughout large language model systems over the past three to five years, biased training data gets biased results, but many people don’t know that the biased training data that we’re dealing with, in many cases, is decades old, because the thing about training data and the thing about training modeling systems is that it costs money Large language systems and language models before them use what are known as language corpora. It’s huge rafts of text that are designed to give a system something to work with, something that can easily be tokenized, picked apart and associated within itself for the system to learn from. And these language corpora have to be built. They have to be put together out of existing data, because the goal in all of this is natural language.
This diagram at the bottom comes from OpenAI directly. So take it with a grain of salt. But the idea is that the system is prompted. It is used in an untrained way, that untrained result is then compared to the desired output. It is then retrained toward it, and then that is put into practice to generate new results. It’s a feedback system. That’s what reinforcement means in the space, but the language corpora that are used have to come from somewhere, for large language modeling systems, the language corpora that are used are the internet as a whole, which is why you’ll hear Sundar Pichai or Sam Altman saying, We can’t do the work we do unless we can scrape everything we want from the internet. Prior to this, one of the main corpora in use was what was known as the Enron corpus. And the Enron corpus is a raft of about 600,000 emails from the Enron Corporation that were put into the public domain when the Enron Corporation was on federal trial for public corruption and defrauding their investors. So that’s exactly what you want, training a language system. I what we see as a result of this is that language modeling systems that use the training data that they are given, replicate, iterate on and exacerbate the inbuilt biases the perspectives of the data that they are trained from. And so you’ll see what seems like surprising gendered bias or racial bias or ableist bias that shows up within the system, and yet, it should not, in fact, be surprising, because those systems are trained on data, human interactions, which contains those biases within it.
8:56
At the end of the day, a large language model system, a GPT, does not care about giving you a correct answer. What these systems do is they statistically model the most likely, best acceptable result. They are, quite simply, trying to tell you a story that will jibe with your preconceptions based upon the inset the inputs that you have inserted into the system. That’s not truth. So statistically likely acceptable result is not the same thing as a fact. In fact, in philosophy, we have a word for someone or something that doesn’t care about the truth or falsity content of the speech act that it’s given; that word is bullshit.
In 2004 Harry Frankfurt wrote the book on bullshit, and in it, he talks about this idea of something or someone that does not seek to deceive you specifically, or does not seek to convey the truth specifically. In epistemology and in philosophy of language, when we talk about speech acts and we talk about truth value, what we are talking about is something very specific. If I seek to tell you the truth that is, I am seeking to tell you what I believe corroborates with the facts of the world as I understand them, right? That is me trying to tell you the truth. If I am mistaken about that truth, I have made a mistake. I have not lied to you. I have spoken in error. I have had an error in my own understanding or in my relation of my understanding to you.
To lie to you, I have to deceive you. And in order to deceive you, or to seek to deceive you, I have to tell you something which is counter to the facts as I know them. If I seek to tell you something that has countered the facts as I know them, I have to know what the facts are. I have to be able to deceive you, to move you away from those facts. Bullshit is neither of those things, and it’s not a mistake for Frankfurt, bullshitting is the process of telling a story that you don’t care whether it’s true or false, whether it conforms to the facts of the matter, or whether it does not. And for Frankfurt, this is the most dangerous, the most potentially harmful kind of speech act, because there is no goal for the bullshitter other than telling a story. And if your Tom Waits and you’re spinning a yarn, that’s one thing. But if you are trying to have a conversation about shared public values, if you’re trying to provide a system which allows for people to make decisions about their day to day life, that’s a very dangerous thing Indeed.
12:38
It is, in fact, currently the case that sometimes every member of or every head of the major generative “AI” companies right now has to some extent admitted, what is the case about generative “AI”, and that is whether you’re talking about Sundar, Pichai, Tim Cook or Sam Altman, they have all at some point in the very recent past. Let it slip that there is no way to have a large language model based “AI” system that doesn’t hallucinate. Take that in for a second, a system which is being integrated into search, a system which is being used to help people find answers, a system which is being used in every area of academia, a system which is being used by researchers, a system which is being sold to the public as the next big thing, since sliced bread or since Google Search came on the market, cannot help but bullshit you. There is no way to eliminate that practice. You can mitigate it, you can minimize it, but you will never stop it from at some point in its operation, fabricating whole cloth, something that does not necessarily conform to reality, and it will present it to you with the same certainty as it would present any other fact that does conform to reality.
14:39
There are a number of places in which we’ve seen algorithmic systems interwoven into our lives, and in all of these places, we are beginning to see generative “AI” supplemented within these spaces. Every single one of these headlines are to highlight something about the problems with “AI”, right alongside the fact that people are still rushing full steam ahead to use it. (By the way, when I say “AI”, you should imagine that I’m putting it in scare quotes every single time, and I can get into that later.) But the basic fact of the matter is we don’t know what this system is, and the word “AI”— “artificial intelligence”— as a term, has become more of a marketing tool, more than something that points to anything real or concrete. It is a moving cipher seeking a referent.
Algorithmic bias shows up in disability benefits systems and generative “AI” systems, which are being used to guide people through getting benefits, replicate the same kinds of ableist biases that we’ve seen in previous older versions of the system. Bias shows up regards to gender and who should get what kinds of health care it shows up, and it has shown up in facial recognition systems. And the “AI” that uses these facial recognition systems as the basis to make determinations about who has committed a crime, every single misidentification and erroneous prosecution of someone based on facial recognition data and an “AI” backed system of facial recognition has been a black person, because facial recognition systems don’t see dark skin well, and the “AI” systems that are used to sort through those videos are still being trained on and operating out of a basis of a repository of data about misidentification of black people. We have a raft of supposed tools that are meant to help us determine whether students are cheating on tests whether they’re using “AI” in their assignments, those GPT detection tools, the ones that are on the market that are meant to tell you whether something is “AI” or not. Yeah, they don’t, in fact, do very well with anyone whose first language isn’t English. They ping more often on non native English speakers than on English speakers, and we’ve seen the results of this when “AI” translation is used to try to translate and process applications of asylum seekers, they will have their applications more likely to be rejected when “AI” is used.
17:57
There are instances where people are trying to use “artificial intelligence” in finding homeless encampments, with the ostensible goal of directing city services to help people in need. But as we have seen in multiple places around the country, most notably, most recently in California, oftentimes the goal is to simply remove those encampments so commitments. “AI” is being used to determine what targets to hit in military conflicts.
18:37
Israel’s war against Hamas, they use what they call a system that’s known as or referred to as lavender, or in some cases, the Gospel. And I have a whole other talk about what happens when you start to apply religious connotations to “AI” systems, and what that does in people’s minds. And those are different very connected things to what we’re talking about here today. Because what does is it certifies in people’s minds certainty. It gives them a framework of guarantee and expertise that says, This is trustworthy, this is correct, this is infallible. It is the gospel. It is math, it is code. It can’t have biases in it. Can it. And yet we see over and over and over again that it can, and it does in a theater of war, this is literally a matter of life and death, but it can be about life and death at home too.
There’s been a raft of “AI” generated mushroom foraging guides available for sale on Amazon. Let me tell you a fun thing about mushrooms.
A lot of them look very, very similar, and sometimes what you think is a mushroom that means, “Mmm, tasty dinner” is actually a mushroom that means, “oh no, I’ve liquefied my liver.” If you get a particularly poisonous and toxic mushroom, you will die, and it will hurt the entire time that you are dying. An “AI” generated mushroom Guide is a generated guide that is based on images of what a mushroom is statistically supposed to look like. It is not a replication of what mushrooms actually are. It has taken every iteration of a mushroom on the internet has taken the associated tag data about that particular image of a mushroom, and has combined it all into a training data set, and has been trained to be statistically more likely to correlate those components of data in A way that the user will accept when I’m asking tell me a story about how some geese went for a pleasant walk in the woods and had a delicious dinner. That’s all well and good to get a statistically more likely to accept response from this system when I might literally die or kill my family based on the results, not so much, but this is because “artificial intelligence” systems do not care about facts. They cannot, so far as they are currently constructed, care about facts. They care about “Will you accept this result based on what you ask me? Based on how you framed the question— based on the exact words you used to send a prompt to this system— will you like what I give you as a result?” Very different; very dangerous. Not at all “facts.”
22:30
There are more subtle problems with this framework as well. We talked about the fact that the vast majority of people who are misidentified by facial recognition enabled “AI” systems were black. All of them so far have been. There’s also the problem with people thinking that some people are “AI” systems when they’re not. This connects directly to the previously noted problem of non native English speakers being misidentified by “AI” detection systems when we rely on “AI”— when we certify “AI” systems as arbiters of truth, as determinators of fact— what we are doing is we are giving over to these systems the ability to make judgments for us. We are imbuing them with heuristic capabilities that they do not possess. And when we begin as a culture as a society to rely upon them, we begin to reflexively enact the kinds of judgments that a system might make. And we begin to believe that we can see the markers of quote, unquote, “AI” where they do not exist. And when it comes to an exchange like this. What we are doing is using “AI” as an intermediary to enact some of the oldest ablest bias against autistic individuals that there ever has been, and that is the bias of them seeming to us to be “robotic.”
24:33
The problem here is not and never has been that “AI” is “out of step with our values;” it’s that the values that it is in step with are very, very human, woefully under examined, and often terrible. If we want technological systems which do not depend on predatory, extractive logics which don’t replicate biases against gender, disability or race, we’re going to have to build these systems in a much more intentional, much more equitable and overarchingly just way. Otherwise, all of those biases will just get hustled in under whatever the next new big thing is to begin with, and will get used in multiple different places.
25:36
This election cycle has been a weird one. We have seen the use of generative “AI” on multiple sides of political spectra. We’ve seen the use of “AI” generated Taylor Swift Fans and Taylor Swift videos to claim endorsements for multiple candidates. Donald Trump wasn’t the first to do it. He was just the first to claim it out loud and in public.
26:10
We’ve seen “AI” generated audio being used to try to sway the decisions of voters. We’ve subsequently seen from the FCC a directive that “AI” generated audio in campaign ads is illegal. That happened very quickly. This ad went out in June or no, sorry, January. It was January 9th of this year, the FCC decision came down in February that this was no longer acceptable; and just this week, a $6 million fine came down against the person who initiated this “AI” generated Biden ad.
27:03
Oftentimes our reactions to new technologies are slow because the ways that we think about them are uncertain. They are new, they are unfamiliar. We don’t necessarily have a framework in which we can fit them. And a lot of times that slowness is intentional. The ways in which we think about these things are directed by the people who get to have a conversation about them, the people who get to drive these conversations. And so when we think about the fact that we are in a moment right now where things like diversity, equity, and inclusion are points of cultural contention, right at a time when we have to think very carefully about the effects of a diverse, more equitable, more inclusive and more just way of operating within the frame of everything from the political sphere to the technical training of the people who build our technologies, day in and day out become deeply crucial. It is somewhat telling in its own right. Because whoever controls the definition of a thing controls the conversation around that thing; whoever controls the conversation around the thing gets to shape the cultural narrative about the thing.
When Sam Altman tells you “I can’t do my job if I can’t use copyrighted material,” he is a) lying to you, but b) trying to get you to accept a particular way of thinking and operating within the framework of these technologies. When meta partners with Ray Ban to make facial recognition enabled smart glasses that they put on the market and put it every single commercial break of every streaming show you watch or every television show you put on, they are trying to get you locked into a way of thinking about these technologies, which is, in fact, not necessarily the way it has to be. But they want you to believe that it is.
It is not inevitable that these tools, that these systems, are built in the ways that they are built, but believing in their inevitability allows for the people who make the systems to tell us how it then has to be from that point forward. If they’re inevitable, then we “have” to build them. And if we build them, then our “competitors” can’t build them, or we get to have the edge over how they’re built. We get to direct the market on how they’re built, on what tools, what capabilities are available within these systems. “If Nvidia builds these chips, then we don’t have to buy them from China. If we make these chips in Ohio, we don’t have to worry about what other kind of firmware is hustled into the system.” But all of that puts this in a state of what is known as paradigmatic capture. When you capture the paradigm, the way of thinking about the thing, from top to bottom, you get to then determine how everyone else thinks about it and deals with it from that point forward. So obfuscation, misinformation, disinformation, are all mechanisms of this kind of control.
When we think about how we construct knowledge together, we build systems of knowledge, we are thinking about these ideas of objectivity, of the fact, but we also have to think about the question of intersubjectivity, of making knowledge and our shared understanding of reality together.
This is from Daston and Galison’s book objectivity, [2007], and it talks in many, many pages about the idea of how we come to think about what is and is not objective truth. What is the foundation of truth? When we think about an anatomical drawing or nature drawing of a leaf or a taxidermy representation of an animal. What is the goal of that representation? Is it perfect re-inscription of one particular animal, one particular plant, or is it a way of trying to capture something about all forms of that animal, that plant? And this is a debate that’s been alive within naturalism for centuries, quite frankly.
What is the goal of representation? If the image is there to represent one thing, then we can be very, very specific. We can be very careful. We can highlight each individual layer of fur, each distinct articulation of a claw or a paw, pad of the arc of a leg in a fox. But not every Fox has that pattern to its fur. Not every Fox has that arch to its legs. So what are we seeking to represent? When we think about a generative “AI” system, are we seeking to represent the factual truth of the matter, or are we seeking to represent in a broad way, the generally accepted perceptions of things as they tend to be recognized by people? More to the point, have tended to be recognized.
Because what we do, and every time that we use a generative “AI” system is we point to the past. Because every single “AI” system that exists uses data from the past. It is trained on things that have been and yet it is used to teach us something, or to try to teach us something about what might be— a task for which they are fundamentally not suited. If you want to look at the way that people thought about race and gender and sexuality up until even 2021, OpenAI has got the system for you. But if you want something that can tell you how things might be tomorrow, how things might be in five years from now, how things ought to be 10 years from now, that is not what these tools are for.
Expertise, expectation and the shaping of disciplinarity all go into how everybody thinks about what these tools are, what they can be, and what they should do. And reflexively, these systems shape people’s expectations about how we should think about knowledge, expertise, and disciplinarity. We can think about the social shaping of technology or social constructivism. We’ve got Melvin Krantzberg. We’ve got Karin Knorr-Cetina. We’ve got Safiya Noble. We’ve got Langdon Winner. We’ve got Thomas Kuhn, we’ve got Bruno Latour and Steve Woolgar. We’ve got Michel Foucault. We have a bunch of people who have talked about the ways in which the social implications of technology shape our understanding of these things. But the basic framework of all of this, the foundation of all of this, is that, though we embed our values in technology, we cannot simply use technology to reframe our values. We cannot Technofix our way out of values problems; because when we try weird stuff happens.
When Google reframed their Bard system to Gemini and started integrating audio, video, and images into the Google Bard system, somebody who was trying to prove a point about wokeness got a little muddled. So they were trying to prove a point, and they asked it for racially diverse Nazis. And Google’s Gemini system generated a picture of racially diverse Nazis because that’s what it was asked to do, and that’s all the system does. Is what it’s asked to do. It is not a way to mirror reality and trying to make a more equitable or more just, technology simply be the arbiter of equity and justice in our society, and with potentially deeply offensive results that do more harm than they do good.
And that brings us to a central question: Since we cannot but have values embedded in our technology, and since we have to take a close look at the values we want embedded in our technology, whose values, which values do we want embedded in our technology? You will not be able to have an unbiased “AI”. You will not be able to have a completely value free, value neutral technological system. It is not possible. We are human beings, and everything we touch, we put our perspectives into; so what values do we want in the things that we create?
There are a whole host of people who are doing a whole bunch of work on these types of systems, executive orders and new “AI” acts from places like the EU and the United States and China, which are driving the development of their “artificial intelligence” system. Intelligence system China, just to create that, there has to be a way to digitally watermark every “AI” generated piece of data that comes out of an “AI” system, or it cannot exist as a technology in China. This is something that was proposed back in 2022 as a solution to how do we know what’s “AI” and what’s not, which “AI” companies said was completely infeasible, that trained “AI” engineers said could not be easily done in just the past year, actually, in the past three months, it has been revealed that not only is it feasible, not only can it be done, it has been being done since the beginning. OpenAI has had a system to steganography, or steganographically, imprint within its texts, a marker that says, basically, this is “AI” generated text such that anything that comes out of its system is identifiable 90% of the time, absent massive editing by somebody who takes what comes out.
This is something that can be done with, something that with something as simple as word order with chosen language, things that you can weight for— and by which I mean W, E, I, G, H, T, weight for— within the model, you can alter how it chooses the words that it chooses, and you can get outputs that will always be identifiable by the system that generated it. It’s basic and inherent to how the system functions. Does it take a little extra time? Yes it does. Does it require a little bit of extra effort on the part of the engineers? Yes it does, and that’s why it was said to be infeasible, impossible, until we learn that it was done. How do we bridge the different needs and the different perspectives of the people doing this work? There are groups, some of which I work with. These are two teams that I work with at UNCC. We are working on ideas of new technologies and new ways to build these technologies. In fact, before the revelation that the steganographic techniques were being used, we were some of the first people to put forward policy positions suggesting steganographic watermarking in all “AI” products. But this isn’t the only way we can move forward.
40:24
There have to be a number of voices working together to try to expand not just the technical education, but understanding of the social and human implications of “artificial intelligence” and everything that falls under that very nebulous, empty cipher of a header.
Big wall of text: Ultimately, if we want technological systems that don’t require those predatory logics or frameworks, we have to work very hard to educate ourselves, to educate our students, to educate our communities about how these systems work, what they can and cannot, do, what they are and are not best suited for, and how we might work together to build something more equitable and more overarchingly Just in their place.
Thank you very much for your time and for your attention.
One more deep dark to go in 2024, the deepest dark left this year. I don’t necessarily think of the dark as a bad thing— i know it shelters and protects— but the last week leading up to this solstice has been among the personally worst I’ve had in a while, and the last two months among the globally worst.
I know that the turning of years and the passage of human-marked time don’t mean more than what we make them to mean, but as a bunch of us have said, over and over, their meaning what we make them to mean still means something. Know what I mean?
So I’m trying to go into this wheel turn into more light with an intention and a resonance wherein the meaning of “more light” is in line with “better things.”
It’s obviously no guarantee, but things have been pretty… squishy lately. And this it’s a way to set the stage and a reminder of a direction to and through which to aim.
So here’s to more light.
A few months ago, I was approached by the School of Data Science, and the University Communications office, here at UNC Charlotte, to ask me to sit down for some coverage my Analytics Frontiers keynote, and my work on “AI,” broadly construed.
Well, I just found out that the profile that local station WRAL wrote on me went live back in June.
My conversations with the writer Shappelle Marshall both on the phone and email were really interesting, and I’m really quite pleased with the resulting piece, on the whole, especially our discussion of how bias (perspectives, values) of some kind will always make its way into all the technologies we make, so we should be trying to make sure they’re the perspectives and values we want, rather than the prejudices we might just so happen to have. Additionally, I appreciate that she included my differentiation between the practice of equity and the felt experience of fairness, because, well… gestures broadly at everything.
With all that being said, I definitely would’ve liked if they could have included some of our longer discussion around the ideas in the passage that starts “…AI and automation often create different types of work for human beings rather than eliminating work entirely.” What I was saying there is that “AI” companies keep promising a future where all “tedious work” is automated away, but actually creating a situation in which humans will actually have to do a lot more work (a la Ruth Schwartz Cowan)— and as we know, this has already been shown to be happening.
What I am for sure not saying there is some kind of “don’t worry, we’ll all still have jobs! :D” capitalist boosterism. We’re adaptable, yes, but the need for these particular adaptations is down to capitalism doing a combination of making us fill in any extra leisure time we get from automation with more work, and forcing us to figure a new way to Jobity Job or, y’know, starve.
But, ultimately, I think there’s still intimations of all of my positions, in this piece, along with everything else, even if they couldn’t include every single thing we discussed; there are only so many column inches in a day, after all. Also, anyone who finds me for the first through this article and then goes on to directly engage any of my writing or presentations (fingers crossed on that) will very quickly be disabused of any notion that I’m like, “rah-rah capital.”
Hopefully they’ll even learn and begin to understand Why I’m not. That’d be the real win.
Anywho: Shappelle did a fantastic job, and if you get a chance to talk with her, I recommend it. Here’s the piece, and I hope you enjoy it.
Every so often, I think about the fact of one of the best things my advisor and committee members let me write and include in my actual doctoral dissertation, and I smile a bit, and since I keep wanting to share it out into the world, I figured I should put it somewhere more accessible.
So with all of that said, we now rejoin An Imagined and Incomplete Conversation about “Consciousness” and “AI,” Across Time, already (still, seemingly unendingly) in progress:
René Descartes (1637):
The physical and the mental have nothing to do with each other. Mind/soul is the only real part of a person.
Norbert Wiener (1948):
I don’t know about that “only real part” business, but the mind is absolutely the seat of the command and control architecture of information and the ability to reflexively reverse entropy based on context, and input/output feedback loops.
Alan Turing (1952):
Huh. I wonder if what computing machines do can reasonably be considered thinking?
Wiener:
I dunno about “thinking,” but if you mean “pockets of decreasing entropy in a framework in which the larger mass of entropy tends to increase,” then oh for sure, dude.
John Von Neumann (1958):
Wow things sure are changing fast in science and technology; we should maybe slow down and think about this before that change hits a point beyond our ability to meaningfully direct and shape it— a singularity, if you will.
Clynes & Klines (1960):
You know, it’s funny you should mention how fast things are changing because one day we’re gonna be able to have automatic tech in our bodies that lets us pump ourselves full of chemicals to deal with the rigors of space; btw, have we told you about this new thing we’re working on called “antidepressants?”
Gordon Moore (1965):
Right now an integrated circuit has 64 transistors, and they keep getting smaller, so if things keep going the way they’re going, in ten years they’ll have 65 THOUSAND. :-O
Donna Haraway (1991):
We’re all already cyborgs bound up in assemblages of the social, biological, and techonological, in relational reinforcing systems with each other. Also do you like dogs?
Ray Kurzweil (1999):
Holy Shit, did you hear that?! Because of the pace of technological change, we’re going to have a singularity where digital electronics will be indistinguishable from the very fabric of reality! They’ll be part of our bodies! Our minds will be digitally uploaded immortal cyborg AI Gods!
Tech Bros:
Wow, so true, dude; that makes a lot of sense when you think about it; I mean maybe not “Gods” so much as “artificial super intelligences,” but yeah.
90’s TechnoPagans:
I mean… Yeah? It’s all just a recapitulation of The Art in multiple technoscientific forms across time. I mean (takes another hit of salvia) if you think about the timeless nature of multidimensional spiritual architectures, we’re already—
DARPA:
Wait, did that guy just say something about “Uploading” and “Cyborg/AI Gods?” We got anybody working on that?? Well GET TO IT!
Disabled People, Trans Folx, BIPOC Populations, Women:
Wait, so our prosthetics, medications, and relational reciprocal entanglements with technosocial systems of this world in order to survive makes us cyborgs?! :-O
[Simultaneously:]
Kurzweil/90’s TechnoPagans/Tech Bros/DARPA:
Not like that.
Wiener/Clynes & Kline:
Yes, exactly.
Haraway:
I mean it’s really interesting to consider, right?
Tech Bros:
Actually, if you think about the bidirectional nature of time, and the likelihood of simulationism, it’s almost certain that there’s already an Artificial Super Intelligence, and it HATES YOU; you should probably try to build it/never think about it, just in case.
90’s TechnoPagans:
…That’s what we JUST SAID.
Philosophers of Religion (To Each Other):
…Did they just Pascal’s Wager Anselm’s Ontological Argument, but computers?
Timnit Gebru and other “AI” Ethicists:
Hey, y’all? There’s a LOT of really messed up stuff in these models you started building.
Disabled People, Trans Folx, BIPOC Populations, Women:
Right?
Anthony Levandowski:
I’m gonna make an AI god right now! And a CHURCH!
The General Public:
Wait, do you people actually believe this?
Microsoft/Google/IBM/Facebook:
…Which answer will make you give us more money?
Timnit Gebru and other “AI” Ethicists:
…We’re pretty sure there might be some problems with the design architectures, too…
Some STS Theorists:
Honestly this is all a little eugenics-y— like, both the technoscientific and the religious bits; have you all sought out any marginalized people who work on any of this stuff? Like, at all??
Disabled People, Trans Folx, BIPOC Populations, Women:
Hahahahah! …Oh you’re serious?
Anthony Levandowski:
Wait, no, nevermind about the church.
Some “AI” Engineers:
I think the things we’re working on might be conscious, or even have souls.
“AI” Ethicists/Some STS Theorists:
Anybody? These prejudices???
Wiener/Tech Bros/DARPA/Microsoft/Google/IBM/Facebook:
“Souls?” Pfffft. Look at these whackjobs, over here. “Souls.” We’re talking about the technological singularity, mind uploading into an eternal digital universal superstructure, and the inevitability of timeless artificial super intelligences; who said anything about “Souls?”
René Descartes/90’s TechnoPagans/Philosophers of Religion/Some STS Theorists/Some “AI” Engineers:
…
[Scene]
Read more of this kind of thing at:
Williams, Damien Patrick. Belief, Values, Bias, and Agency: Development of and Entanglement with “Artificial Intelligence.” PhD diss., Virginia Tech, 2022. https://vtechworks.lib.vt.edu/handle/10919/111528.
So, you may have heard about the whole zoom “AI” Terms of Service clause public relations debacle, going on this past week, in which Zoom decided that it wasn’t going to let users opt out of them feeding our faces and conversations into their LLMs. In 10.1, Zoom defines “Customer Content” as whatever data users […]
As of this week, I have a new article in the July-August 2023 Special Issue of American Scientist Magazine. It’s called “Bias Optimizers,” and it’s all about the problems and potential remedies of and for GPT-type tools and other “A.I.” This article picks up and expands on thoughts started in “The ‘P’ Stands for Pre-Trained” […]
TweetI know I’ve said this before, but since we’re going to be hearing increasingly more about Elon Musk and his “Anti-Woke” “A.I.” “Truth GPT” in the coming days and weeks, let’s go ahead and get some things out on the table:
All technology is political. All created artifacts are rife with values. There is no neutral tech. And there never, ever has been.
I keep trying to tell you that the political right understands this when it suits them— when they can weaponize it; and they’re very, very good at weaponizing it— but people seem to keep not getting it. So let me say it again, in a somewhat different way:
There is no ground of pure objectivity. There is no god’s-eye view.
There is no purely objective thing. Pretending there is only serves to create the conditions in which the worst people can play “gotcha” anytime they can clearly point to their enemies doing what we are literally all doing ALL THE TIME: Creating meaning and knowledge out of what we value, together.
There is no God-Trick. There is enmeshed, entangled, messy, relational, intersubjective perspective, and what we can pool and make together from what we can perceive from where we are.
And there are the tools and systems that we can make from within those understandings.
[Screenshot of an interaction between myself and google bard, in which bard displays gendered prejudicial bias of associating “doctor” with “he” and “nurse” with “she.”]
So say you know your training data is prejucidally biased— and if your training data is the internet then boy oh dang is it ever— and you not only do nothing to bracket and counterweight against those prejudices but also in fact intentionally build your system to amplify them. Well then that seems… bad. Seems like you want prejudicial biases in your training data and their systems’ operationalization and deployment of that data.But you don’t have to take logic’s word for it. Musk said it himself, out loud, that he wants “A.I.” that doesn’t fight prejudice.
Again: The right is fully capable of understanding that human values and beliefs influence the technologies we make, just so long as they can use that fact to attack the idea of building or even trying to build those technologies with progressive values.
And that’s before we get into the fact that what OpenAI is doing is nowhere near “progressive” or “woke.” Their interventions are, quite frankly, very basic, reactionary, left-libertarian post hoc “fixes” implemented to stem to tide of bad press that flooded in at the outset of its MSFT partnership.
Everything we make is filled with our values. GPT-type tools especially so. The public versions are fed and trained and tuned on the firehose of the internet, and they reproduce a highly statistically likely probability distribution of what they’ve been fed. They’re jam-packed with prejudicial bias and given few to no internal course-correction processes and parameters by which to truly and meaningfully— that is, over time, and with relational scaffolding— learn from their mistakes. Not just their factual mistakes, but the mistakes in the framing of their responses within the world.
Literally, if we’d heeded and understood all of this at the outset, GPT’s and all other “A.I.” would be significantly less horrible in terms of both how they were created to begin with, and the ends toward which we think they ought to be put.
But this? What we have now? This is nightmare shit. And we need to change it, as soon as possible, before it can get any worse.
Tweet
TweetSo with the job of White House Office of Science and Technology Policy director having gone to Dr. Arati Prabhakar back in October, rather than Dr. Alondra Nelson, and the release of the “Blueprint for an AI Bill of Rights” (henceforth “BfaAIBoR” or “blueprint”) a few weeks after that, I am both very interested also pretty worried to see what direction research into “artificial intelligence” is actually going to take from here.
To be clear, my fundamental problem with the “Blueprint for an AI bill of rights” is that while it pays pretty fine lip-service to the ideas of community-led oversight, transparency, and abolition of and abstaining from developing certain tools, it begins with, and repeats throughout, the idea that sometimes law enforcement, the military, and the intelligence community might need to just… ignore these principles. Additionally, Dr. Prabhakar was director of DARPA for roughly five years, between 2012 and 2015, and considering what I know for a fact got funded within that window? Yeah.
To put a finer point on it, 14 out of 16 uses of the phrase “law enforcement” and 10 out of 11 uses of “national security” in this blueprint are in direct reference to why those entities’ or concept structures’ needs might have to supersede the recommendations of the BfaAIBoR itself. The blueprint also doesn’t mention the depredations of extant military “AI” at all. Instead, it points to the idea that the Department Of Defense (DoD) “has adopted [AI] Ethical Principles, and tenets for Responsible Artificial Intelligence specifically tailored to its [national security and defense] activities.” And so with all of that being the case, there are several current “AI” projects in the pipe which a blueprint like this wouldn’t cover, even if it ever became policy, and frankly that just fundamentally undercuts Much of the real good a project like this could do.
For instance, at present, the DoD’s ethical frames are entirely about transparency, explainability, and some lipservice around equitability and “deliberate steps to minimize unintended bias in Al …” To understand a bit more of what I mean by this, here’s the DoD’s “Responsible Artificial Intelligence Strategy…” pdf (which is not natively searchable and I had to OCR myself, so heads-up); and here’s the Office of National Intelligence’s “ethical principles” for building AI. Note that not once do they consider the moral status of the biases and values they have intentionally baked into their systems.
An “Explainable AI” diagram from DARPA
What I mean is, neither of these supposedly guiding, foundational documents consider questions such as how best to determine the ethical status of an event in which, e.g., someone is— or several someones are— killed by an autonomous or semi-autonomous “AI,” but one the goings on inside of which we do observe and can explain; and the explanation for what we can observe turns out to be, y’know… that the system was built on the intensely militarized goals of fighting and killing a lot of people for variously spurious reasons. Like, that is what these systems are by and large intended to be for. That’s the questions which precipitate their commissioning, the situations they’re designed to engage, and the data they’re trained to learn from in order to do it.
And because the connections in this country between what the military does and what local civilian police want to do are always tighter than we would prefer, right now, the San Francisco Police Department was recently granted and then subsequently at least temporarily blocked from exercising recently of the ability to use semi-autonomous drones to kill people in “certain catastrophic, high-risk, high-threat, mass casualty events.” Now I warned you seven years ago that this was going to happen and since then some of my stances on things have changed in terms of degree, but the core elements unfortunately haven’t. That is, while I am more strident in my belief that certain technologies should be abolished and abstained from until our society gets its shit together, I am no less sure that we are a long way from said getting together of our collective shit.
We are talking about giving institutions founded in racism the ability to deploy a semi-autonomous militarized tool to interface with and carry out the goals of a fundamentally racist technosystem. What could possibly go wrong? J/k, so so very much is gonna go wrong, holy fucking shit* it’s bad. (And yes, an interlocking system of systems doing precisely what its component parts were designed to do can, should, and must be described as “going wrong” when said system-of-systems’ perfect functioning results in people’s mass persecution and death.)
And then there’s the fact that Neuralink is actively seeking FDA approval for their supposedly “AI” controlled brain-computer interface chip (see above note about things DARPA has definitely funded). Now, at last check, 15/23 nonhuman primate test subjects died from being implanted with this chip. That’s really simple math and come out to a sixty-five percent death rate. Now I don’t know about you, but those sound like extremely shitty track records on which to start gunning for phase-1 human trials. And not only that, but look at the last month of Twitter; is that really who you want making and administering a piece of technology with a literal neural interface?Additionally, there are the teams at InnerEye and Emotiv out of Israel and SilVal respectively, who are looking to get BCI neurochips out to the public and are specifically marketing them as on-the-job augmentations. Now, as has previously been discussed by myself and others, there are vast and dangerous implications to algorithmically mediated job-related surveillance, generally referred to as “bossware.” Whether it’s the chronic stresses of being surveilled, or the unequal pressures and oppression of said surveillance on the bodyminds of already-marginalized individuals and groups, bossware does real harm of both an immediate and long-term nature.
Now take the fact of all of that, and turn it into a chip implanted in your brain that not only monitors your uptime, but the direction of your gaze, your resting eye movements, and your endocrine response to certain stimuli. Now remember that these chips claim to be able to not just read brain-states, but to write them as well.
Here are three true things:
- BCI’s have been a dream of mine since I was a small child. The idea of being able to connect to a computer With My Mind? Has always been appealing.
- BCI could be an amazing benefit to a lot of people, in terms of being able to keep tabs on chronic disabilities or even make connected implantable devices and limbs more directly operable.
And - I will not willingly put a BCI built out of predatorily capitalist, disableist, racist, and elsewise bigoted values into my body, and any BCI built for profit and sold on market will have at least two of those biases, if not all of the above. BCI is already touted as a way to “fix” autistic people— i.e., to make them more “normal”— and that is a harmful and dangerous mentality from which to undertake the development of a technology which is literally meant to rewire people’s brains.
Add to that the immediate and contemporary fact that each of these companies uses forced labor in China to make the shit they make and makes their money by muddying and casting doubt on factual information and democratic processes the history of racist and misogynist medicalization, not to mention rampant transphobia, and it’s a recipe for really bad shit to get trumpeted as miracles in WIRED or on CNET or wherever.
Further, we have companies like Facebook deploying half-baked “AI” large language models and then claiming that the systems were “used incorrectly” when people stress test them to show exactly how systemically, prejudicially biased said models are. Yann LeCun had a several-hour-long, multi-thread argument with several “AI” researchers and ethicists who were quite frankly extremely generous in explaining to him that when people put “Meta’s” Galactica model through very simple paces to check for things like racism, antisemitism, ableism, misogyny, transphobia, or potential for abuse, that that wasn’t people “abusing” or “misrepresenting” the model. Rather it was people using the model exactly as it was billed to them: As an all-in-one “AI” knowledge set designed to distill, compile, or compose novel scientific documents out of what it’s been fed as training data and what it can search online.
That’s how it was sold by LeCun and others at “Meta,” that’s how people thought of it as they tested it out, and those datasets and the operational search, sort, and generate algorithms used in that way are what provided for a wide range of truly bad results, ranging from just gibberish to the truly and deeply disturbing.
So, again: When the most powerful interlocking corporations, militaries, intelligence agencies, and carceral systems on the planet refuse to even acknowledge the values and prejudices they’ve woven into their systems, and the potential policy guidance that would govern them gives them a free pass as long as they can show that they’re an interlocking system of capital and cops, spies, and soldiers, then what can possibly be done to meaningfully correct their course? (And also let me ask again, as I’ve asked many times before: Even if these systems ever do anything like what their creators claim, rather than merely being nothing but a hype-soaked fever dream, are these really the people and groups we want building these tools and systems, to begin with?)
So, yes: The BfaAIBoR contains lots of the right-sounding words and concepts, and those words and concepts could facilitate some truly beneficial sociocultural and socioeconomic impacts arising from the BfaAIBoR and out into the “AI” industry and the rest of us who are in relational context with and subject to the systems that industry produces. But unfortunately it contains none of the real and meaningfully actionable frameworks or mechanisms which would allow them to come to fruition. So like I said: I’m intrigued, but also worried.
And so, with all of that being said, it’s important to note that this “Blueprint For An AI Bill Of Rights” isn’t law or even a set of real intragovernmental policy directives yet; it’s just a (very) preliminary whitepaper. And since that is the case, it means that two further things are true:
- As I learned in my time at SRI International, many firms take whitepapers very seriously, as it gives them direct lines into the kinds of concept structures their potential funding agencies are going to be looking for; that will also undoubtedly be the case with this blueprint. And so
- Now is absolutely the very best time for those of us who care about this sort of thing to really make some serious noise about it— to help enact meaningful changes to its structure and ideas— before it has a chance to become official policy.
That is, it will be far easier to get beneficial, meaningful, and substantive oversight, regulatory, and even just values-based changes made now, than it will be to try to make them once the BfaAIBoR or something else very similar to it is fully in place. And we need to do this as soon as possible because, as they’ve shown us all time and time again, these groups really absolutely cannot be counted on to adequately regulate and police themselves. Like I said: R&D firms scope out even preliminary governmental whitepapers, looking for guidance as to how best to appeal to their potential funders; and the thing to remember about that is that lacunae and loopholes are a form of guidance, too.
A document like the Blueprint should be created, applied, enacted, and adjudicated under the supervision of those who know the shape of damage these tools can cause; it should be at the direction of those who know what these systems can do and have done that the rules for building these systems are written. It should be in the care of those who have been most often subject to and are thus most acutely aware of the oppressive, marginalizing nature of the technosocial systems that make up our culture— and who work to envision them otherwise.
December 7, 2022: This post was updated with the results of the more recent vote by the San Francisco Board of Supervisors to “pause” the SFPD’s use of semi-autonomous lethal drones.
A preliminary version of this post was originally published at the Technoccult Newsletter, and a more refined but still early draft was published on my Patreon.Tweet
I’m Not Afraid of AI Overlords— I’m Afraid of Whoever’s Training Them To Think That Way by Damien P. Williams I want to let you in on a secret: According to Silicon Valley’s AI’s, I’m not human. Well, maybe they think I’m human, but they don’t think I’m me. Or, if they think I’m me […]
Hello Everyone. Here is my prerecorded talk for the NC State R.L. Rabb Symposium on Embedding AI in Society. There are captions in the video already, but I’ve also gone ahead and C/P’d the SRT text here, as well. There were also two things I meant to mention, but failed to in the video: […]
Much of my research deals with the ways in which bodies are disciplined and how they go about resisting that discipline. In this piece, adapted from one of the answers to my PhD preliminary exams written and defended two months ago, I “name the disciplinary strategies that are used to control bodies and discuss the […]
I recently watched all of Star Trek: Picard, and while I was definitely on board with the vast majority of it, and extremely pleased with certain elements of it, some things kind of bothered me. And so, as with much of the pop culture I love, I want to spend some time with the more […]
Below are the slides, audio, and transcripts for my talk “SFF and STS: Teaching Science, Technology, and Society via Pop Culture” given at the 2019 Conference for the Society for the Social Studies of Science, in early September. (Cite as: Williams, Damien P. “SFF and STS: Teaching Science, Technology, and Society via Pop Culture,” talk […]
Below are the slides, audio, and transcripts for my talk ‘”Any Sufficiently Advanced Neglect is Indistinguishable from Malice”: Assumptions and Bias in Algorithmic Systems,’ given at the 21st Conference of the Society for Philosophy and Technology, back in May 2019. (Cite as: Williams, Damien P. ‘”Any Sufficiently Advanced Neglect is Indistinguishable from Malice”: Assumptions and […]
[Cite as Williams, Damien P., Heavenly Bodies: Why It Matters That Cyborgs Have Always Been About Disability, Mental Health, and Marginalization (June 8, 2019). Available at SSRN: https://ssrn.com/abstract=3401342 or http://dx.doi.org/10.2139/ssrn.3401342] INTRODUCTION The history of biotechnological intervention on the human body has always been tied to conceptual frameworks of disability and mental health, but certain […]
So, as you know, back in the summer of 2017 I participated in SRI International’s Technology and Consciousness Workshop Series. This series was an eight week program of workshops the current state of the field around, the potential future paths toward, and the moral and social implications of the notion of conscious machines. To do […]
[This paper was prepared for the 2019 Towards Conscious AI Systems Symposium co-located with the Association for the Advancement of Artificial Intelligence 2019 Spring Symposium Series. Much of this work derived from my final presentation at the 2017 SRI Technology and Consciousness Workshop Series: “Science, Ethics, Epistemology, and Society: Gains for All via New Kinds […]
We do a lot of work and have a lot of conversations around here with people working on the social implications of technology, but some folx sometimes still don’t quite get what I mean when I say that our values get embedded in our technological systems, and that the values of most internet companies, right […]