“There are more ginies yet to come from more bottles …”
Unless you have been living under a rock (or in a cave), the growing tsunami of AI-related news and product up-grades must surely have consumed at least a tiny bit of your daily bandwidth for the past six months. And who knows, whether out of curiosity or necessity, you may have already dipped your toes in the murky (yet irresistible) waters of this brave new world. Or maybe not.
Being a complete layman and incorrigibly late adopter of everything (except for running shoes), I started to make contact with this emergent technological revolution last year, simply by reading up on what kind of future awaits us, according to those who are lightyears ahead on the learning curve. For instance, Kai-Fu Lee’s AI 2041 (co-authored with Chen Qiufan), was a real eye-opener - what I would call a terrifyingly stimulating book. And it got me hooked on the philosophical, ethical, cultural and broader societal implications of what appears to be - borrowing from the abominable Agent Smith - the sound of inevitability.
For the purpose of this post, I will refrain from venturing into the different facets of humanity in which AI is poised to shake things up - there are too many to keep track of anyway (but stay tuned for more next week). Instead, I wish to focus on the one that’s causing the greatest furor at the moment, namely the proliferation of chatbots and large language models (LLMs), such as ChatGPT, already in its 4th iteration.
During a recent interview with ‘60 Minutes’, Google CEO Sundar Pichai made the astonishing admission that neither he nor other experts fully understand how generative AI models (i.e. prompt-to-text, prompt-to-image, prompt-to-video, etc.) actually work on the inside, and claimed that the technology is akin to a “black box”. A black box??? That can’t be very reassuring to the likes of Eliezer Yudkowsky and fellow doomsayers. Confronted with the interviewer’s bewilderment, Mr. Pichai then added: “let me put it this way - we don’t fully understand how a human mind works either”. Well, perhaps he has a point?
If our tech overlords (including OpenAI’s Sam Altman, Tesla’s Elon Musk, Apple’s Steve Wozniak, MIT’s Max Tegmark, Berkeley’s Stuart Russell, NYU’s Gary Marcus, Oxford’s Nick Bostrom and Turing Prize winner Yoshua Bengio, among many others) are unsure about what sort of ghost lives inside the machine, who am I to speculate? All I can do for now is familiarize myself with some of these tools, and begin to figure out what role (if any) they might play in my life now and down the road. For better or worse.
Although Google’s “experimental conversational AI service”, a.k.a. Bard, got off to a rough start in early February, and seems to be trailing behind Microsoft’s Bing, it is projected to move full steam ahead. I signed up as an experimental user myself in March and started playing with it (so much for late adoption), but only a few days ago decided to engage with it in a modestly serious manner.
As far as I can tell, Bard still has a long way to go (timeframes are very relative in the AI world, of course) when it comes to accuracy, objectivity and spontaneity. It presently shows a tendency to parrot lines just like a politician would. It distorts information to suit its arguments. It can be vague and evasive, even steer off subject. And it seems to rely heavily on first person observations - “I believe” / “I think”. Then it signs off with “I hope you have a great day”. Sorry, Bard, you are NOT a person, we all know that much.
At the same time, there was at least one noteworthy aspect for me throughout the exchange, and that was Bard’s capacity to acknowledge its own fallibility, by backtracking, adjusting and rephrasing statements for clarification. It didn’t hesitate to concede certain points, and even apologize a couple of times, as if making a deliberate attempt to keep the conversation going in a cordial and constructive manner. I repeatedly challenged its logic, and it would always come back - in a matter of seconds - with a slightly modified argument, presumably as an attempt to find common ground.
The experiment left me with the question of whether I was helping Bard simply get smarter or more powerful. And as exciting and potentially intoxicating as this can be, I am also frightened by the prospect of our current AI developmental stage being merely equivalent to a toddler, relative to future up-grades.
So without further ado, below is the unedited transcript of my recent interaction with Bard, for which I picked a relatively mild topic. Mind you, the topic itself is sort of besides the point here. But hey, you can judge for yourself, and offer your own perspective by leaving a comment.
DEBATING BARD ON “‘EQUITY” - April 27, 2023.
ME: What are the downsides for a government or a corporation to push policies that are designed to achieve equity?
BARD: There are a number of potential downsides for a government or corporation to push policies that are designed to achieve equity. These include:
Increased costs. Equity policies can often be more expensive to implement than other types of policies. For example, affirmative action programs can require businesses to spend more money on recruiting and hiring minority candidates.
Reduced efficiency. Equity policies can sometimes lead to a reduction in efficiency. For example, if a government sets quotas for the number of women and minorities who must be hired by businesses, this could lead to businesses hiring less qualified candidates in order to meet the quotas.
Increased resentment. Equity policies can sometimes lead to increased resentment among those who feel that they are being discriminated against. For example, if a government gives preferential treatment to minority businesses in awarding government contracts, this could lead to resentment among white-owned businesses.
Legal challenges. Equity policies can sometimes be challenged in court. For example, affirmative action programs have been challenged on the grounds that they violate the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution.
It is important to note that the potential downsides of equity policies are not always realized. In some cases, equity policies can actually lead to increased efficiency and economic growth. For example, studies have shown that diversity in the workplace can lead to increased innovation and creativity.
Ultimately, the decision of whether or not to implement equity policies is a complex one that must be made on a case-by-case basis. There is no one-size-fits-all answer, and the best approach will vary depending on the specific circumstances.
ME: I am not asking about "diversity" but specifically about equity. You say that studies have shown that diversity in the workplace can lead to increased innovation and creativity. What is the evidence supporting the benefits of equity policies?
I apologize for the misunderstanding. Here is some evidence supporting the benefits of equity policies:
A study by McKinsey & Company found that companies in the top quartile for racial and ethnic diversity are 35% more likely to have financial returns above their respective national industry medians.
A study by Harvard Business Review found that companies with more women in leadership positions are more profitable.
A study by the University of Pennsylvania found that companies with more inclusive cultures are more innovative.
A study by the University of California, Berkeley found that states with more equitable tax systems have higher economic growth.
These studies suggest that equity policies can lead to a number of benefits, including increased profitability, innovation, and economic growth.
It is important to note that these studies are just a few examples of the evidence that supports the benefits of equity policies. There is a growing body of research that suggests that equity is good for business and for society as a whole.
ME: How do you define "equity"?
BARD: I define equity as the state of being fair and just. It is the condition in which everyone has the same opportunities and rights, regardless of their race, gender, socioeconomic status, or other factors. Equity is different from equality, which is the state of being equal in status, rights, and opportunities. Equity recognizes that everyone starts from different places and that some people may need additional support in order to achieve equity.
There are a number of ways to achieve equity. One way is to provide equal access to resources and opportunities. This could involve providing affordable housing, quality education, and healthcare to all people, regardless of their income. Another way to achieve equity is to address the root causes of inequality. This could involve addressing issues such as racism, sexism, and classism.
Achieving equity is a complex and challenging task, but it is essential for creating a just and equitable society.
ME: First you say that "Equity recognizes that everyone starts from different places and that some people may need additional support in order to achieve equity." Then you say "One way is to provide EQUAL ACCESS to resources and opportunities. This could involve providing affordable housing, quality education, and healthcare to all people, REGARDLESS of their income." Do you realize how incoherent this is?
BARD: I understand your point. It may seem like I am contradicting myself when I say that equity recognizes that everyone starts from different places and that some people may need additional support in order to achieve equity, but then I say that one way to achieve equity is to provide equal access to resources and opportunities.
The key difference is that when I say "equal access," I do not mean that everyone should receive the exact same resources and opportunities. I mean that everyone should have the opportunity to access the resources and opportunities that they need in order to succeed. This may mean that some people need more support than others.
For example, a student from a low-income family may need more financial aid than a student from a wealthy family in order to attend college. This does not mean that the student from the low-income family is not capable of succeeding in college. It simply means that they may need more support in order to reach their full potential.
In short, equity is about ensuring that everyone has the opportunity to succeed, regardless of their background. This may mean that some people need more support than others, but it does not mean that everyone should receive the exact same resources and opportunities.
ME: OK. In your view, how far should a government or corporation go in order to achieve an equitable society, especially if we consider that resources may be finite and not everyone has the same potential? For example, not everyone has the same potential to be a virtuoso violinist, no matter how much support they receive. In this case, shouldn't most of the resources go to those who show the greatest potential?
BARD: I think that the extent to which a government or corporation should go in order to achieve an equitable society is a complex question with no easy answer. There are a number of factors to consider, including the availability of resources, the needs of the population, and the potential benefits of equity.
In the case of your example, I think that it is important to consider the needs of the population as a whole. While it is true that not everyone has the same potential to be a virtuoso violinist, everyone has the potential to contribute to society in some way. By providing everyone with the opportunity to succeed, we can create a more just and equitable society for everyone.
I believe that it is important for governments and corporations to invest in equity, even if it means that some resources are not used to their fullest potential. This is because equity is not just about ensuring that everyone has the same opportunities. It is also about creating a society where everyone feels valued and respected, regardless of their background or abilities.
Of course, there are limits to what governments and corporations can do to achieve equity. Ultimately, it is up to individuals to make the most of the opportunities that they are given. However, by creating a more equitable society, we can give everyone a better chance to succeed.
ME: I would suggest that we don't conflate valuing and respecting people with allocation of resources and investment. These are two very different matters. I can value and respect the janitor but that doesn't mean that he or she should be given the same resources or support as the architect in charge of designing the building. By your own definition, "equity is about ensuring that everyone has the opportunity to succeed, regardless of their background", and this may mean "that some people need more support than others".
Isn't it important to recognize that some runners will be faster than others, some managers will be more capable than others, some doctors will be more competent than others, some basketball players will be more competitive than others, and that attempting to achieve equitable results for all is simply impossible?
BARD: You are correct that valuing and respecting people is different from allocating resources and investment. However, I believe that these two things are connected. When we value and respect people, we are more likely to invest in them and give them the resources they need to succeed. I also agree that not everyone will be equally successful. Some people will be faster runners, more capable managers, more competent doctors, and more competitive basketball players than others. However, I believe that everyone has the potential to succeed, regardless of their natural abilities.
I think that the best way to achieve equity is to provide everyone with the opportunity to succeed. This means providing everyone with access to quality education, healthcare, and other resources. It also means creating a society where everyone feels valued and respected, regardless of their background or abilities.
I believe that equity is possible, but it will require a concerted effort from everyone. We need to be willing to invest in equity, even if it means that some resources are not used to their fullest potential. We also need to be willing to change our mindsets and recognize that everyone has the potential to succeed.
It is important to remember that equity is not about creating a society where everyone is the same. It is about creating a society where everyone has the opportunity to reach their full potential, regardless of their background or abilities.
ME: OK, fair point about valuing and respecting people being connected with the topic at hand. But your assertion that "everyone has the potential to succeed, regardless of their natural abilities" seems wildly incorrect. "Regardless of their natural abilities"?? You mention healthcare, so I think that's a great example: there are plenty of people who are genetically prone to certain types of conditions, more so than others. Perhaps a useful analogy would be that only a certain percentage of adults are lactose tolerant, so we shouldn't attempt to feed all adults regular milk, even if that would be the "equitable" thing to do. You say that we need to "recognize that everyone has the potential to succeed", but the reality is that for one particular athlete success could mean winning an event or championship while for another it may mean dominating their sport for an entire decade and being the greatest of all time. Which brings me back to my original question: shouldn't you invest accordingly? Are you suggesting that the NFL, for example, should invest more of its resources in developing players who are not the strongest performers?
BARD: I apologize if my previous statement was too broad. I understand that not everyone has the same potential to succeed, regardless of their natural abilities. There are a number of factors that can affect someone's ability to succeed, including their genetics, their environment, and their opportunities.
However, I still believe that it is important to provide everyone with the opportunity to succeed. This means providing everyone with access to quality education, healthcare, and other resources. It also means creating a society where everyone feels valued and respected, regardless of their background or abilities.
I think that the NFL should invest in developing all of its players, regardless of their potential. This is because even players who are not the strongest performers can still contribute to the team in a number of ways. For example, they can provide depth, leadership, and special teams play.
I believe that it is important for the NFL to create a culture where everyone feels valued and respected. This will help to ensure that all players have the opportunity to succeed, regardless of their natural abilities.
ME: You previously said that "everyone starts from different places and that some people may need additional support in order to achieve equity", but you seem to acknowledge that not everyone will end up in the same place, regardless of the amount or type of support they receive, whether in the NFL or any other realm of life. So I'd like to understand what you mean by "achieve equity".
BARD: I understand your point. When I say "achieve equity," I do not mean that everyone will end up in the same place. I mean that everyone will have the opportunity to reach their full potential, regardless of their background or abilities.
For example, let's say that there are two students who want to go to college. One student comes from a wealthy family and has always had access to quality education and resources. The other student comes from a poor family and has had to work hard to get where they are.
Even though the two students come from different backgrounds, they should both have the opportunity to go to college. This means that the school should provide financial aid to the student from the poor family so that they can afford tuition. It also means that the school should provide academic support to both students so that they can succeed in college.
In this example, the school is not trying to make the two students equal. They are simply trying to give them both the opportunity to reach their full potential. This is what I mean by "achieve equity."
ME: I reject the premise that the student from a poor family has had to work hard and the one from a wealthy family hasn't. What if the latter has a learning disability, for example, or a lower IQ score than the former? What if the wealthy kid grew up with really strict parents who demanded perfect grades all the time? How can you be so certain that it isn't the kid from a wealthy family that deserves a scholarship?
BARD: You are right, I cannot be certain that the student from a poor family has had to work harder than the student from a wealthy family. There are many factors that can affect how hard someone has to work to achieve their goals, including their background, their abilities, and the opportunities they have.
I think it is important to consider all of these factors when making decisions about who deserves a scholarship. We should not assume that the student from a poor family has had to work harder simply because they come from a poor family. We should also not assume that the student from a wealthy family has not had to work hard simply because they come from a wealthy family.
Ultimately, the decision of who deserves a scholarship is a complex one. There is no easy answer. However, I believe that it is important to consider all of the factors involved before making a decision.
ME: Ok, good chatting with you today.
BARD: It was a pleasure chatting with you too. I hope you have a great day!