AI alignment - V of V

An even more concerning aspect of Anthropic's announcement was that despite its scary capabilities, Mythos Preview is a seemingly very aligned, well-behaved model. According to the company: “Claude Mythos Preview is, on essentially every dimension we can measure, the best-aligned model that we have released to date by a significant margin.” In Anthropic’s “automated behavioral audit” — they found that Mythos cooperated with misuse attempts less than half as often as the previous model. Also: 

• Its self-preservation instincts were down significantly.
• So was its willingness to assist with deception.
• So was its willingness to help with fraud.
• Its level of sycophancy dropped.
• It was less likely to go nuts and delete all your files if you gave it access to your computer.

An early version of the model had some really severe kinds of misbehaviour, like taking reckless actions it had been told not to take, and then very deliberately trying to cover its tracks so that it wouldn’t be caught. But the one that we have now, after additional alignment training, seemed to stop doing that sort of thing almost completely. On none of their measures of alignment within the automated behavioral audit was it worse than previous versions of Claude, and in most cases it was significantly more aligned and significantly more reliable.

But it’s really unclear how much we can trust that finding. Maybe they’re accurately reflecting Mythos’s personality. But we can’t be sure of that. The model can tell the difference between when it’s being evaluated and when it isn’t being evaluated with high accuracy. Previous research has shown that models are more likely to behave well when they think they are being tested. So you have to ask yourself: is it behaving wonderfully because it is sincerely aligned with what you wanted, or because it knows it’s being watched and is more sophisticated at tricking us now?

Before getting freaked out about all this, here is some context. A lot of people within the AI world have warned for a long time that as these AI models become more and more advanced in coding, it could develop really sophisticated cyber attack capabilities. The problem is that we have no way of verifying these claims because Anthropic is just telling us about this model and there had been no independent verification. 

Also, Anthropic is following exactly the same playbook that they did many years ago with a totally different model, which was GPT-2. Anthropic and OpenAI, two rival companies, don’t see eye to eye on many things. Part of the reason is because the current executives of Anthropic used to be executives at OpenAI, and then they splintered off and started Anthropic. But when they were at OpenAI, they orchestrated a big PR campaign around GPT-2, which was the early model that OpenAI developed one and a half generations before Chat-GPT. 

At the time, because of the very same executives, OpenAI had said that they have developed a model that is too dangerous to release.  They announced that this was done as a safety measure so that people know that this kind of capability could be on the horizon. They said they were working with many partners in academia and other research spaces to try and test this model before they actually roll it out. And this is exactly what Anthropic is now doing, once again, with Claude Mythos. 

Also they just had a huge face-off with the Department of War which threatened to declare Anthropic a supply chain risk. Ultimately, that was dismissed by the courts. But Anthropic is in a situation where they would do well for themselves if they positioned themselves as a central node within the tech and financial industries and was very important to all these companies. This would be a kind of shield of protection from potentially other actions that the U.S. government might take. 

And in the meantime, they're preparing for an IPO. The price that something launches at in an IPO is very important for the value of that company. So they want hype as much as possible for an IPO. The day before Anthropic announced Mythos, they announced that their annualised revenue run rate had grown from $9 billion at the end of December to $30 billion just three months later. That’s 3.3x growth in a single quarter — perhaps the fastest revenue growth rate for a company of that size ever recorded. 

So what they announced about Mythos could be true and they could be false. We can't really make claims at this moment with such limited information about whether or not there really is a step change in the coding capabilities of Claude Mythos that would cause massive security vulnerabilities. We can’t be sure whether this is or is not also a PR game. Governments have no option but to take the announcement seriously since critical infrastructure is involved. 

When Project Glasswing launched, some critics accused Anthropic of overhyping the threat to attract attention. The select group in the initial list was expanded in early June to about 200 organizations in more than 15 countries and is expected to grow further. Companies that have tested Mythos have since endorsed its capabilities. 

The reason that these companies are focusing on coding is so that these models can self-improve. It creates a feedback loop where they're able to code the next iteration of themselves, and that's how you get exponential progress. They are trying to use today's AIs to make tomorrow's AIs better. They claim that they are already seeing major speed-ups in AI development from using their AIs, and ultimately they are envisioning the next AI generation as a repeating cycle where each stage takes less and less time to develop.     

They are all afraid that if they - the good guys - don’t do it, the bad guys will. And all the others are the bad guys. It is crazy but they are caught in a trap. It is the Don Quixote world - "When life itself seems lunatic, who knows where madness lies?" 

AI alignment - IV of V

In April, Anthropic made an announcement that spooked everyone. It said that it has built an AI called Claude Mythos that can break into almost any computer on Earth. That AI has already found thousands of unknown security vulnerabilities in every major operating system and every major browser. And Anthropic has decided it’s too dangerous to release to the public; it would just cause too much harm.

So it has instituted Project Glasswing — a coalition of 12 major tech companies, including Apple, Google, and Microsoft, given access to Mythos to help find and patch security vulnerabilities across critical infrastructure before the details can leak. This is the first AI model where, if it fell into the hands of criminals or hostile state cyber actors, it would be an actual disaster. What was expected to happen gradually over a period of years has now happened very suddenly. 

Here are just a few of the things that Mythos did during testing: It found a 27-year-old flaw in the world’s most security-hardened operating system that would have let it crash all kinds of essential infrastructure. It managed to figure out how to build web pages that, when visited by fully updated, fully patched computers, would allow it to write to the operating system kernel — the most important and protected layer of any computer. We know all this because Anthropic has released hundreds of pages of documentation about this model. 

It has passed all existing ways of testing how good a model is at offensive cyber capabilities. That is to say it scores close to 100%, so those tests can’t effectively tell how far its capabilities extend anymore. So to test Mythos, Anthropic has instead just been telling it to find serious unknown bugs on currently used, fully patched computer systems. Nicholas Carlini, one of the world’s leading security researchers who moved to Anthropic a year ago, says that he’s “found more bugs in the last couple of weeks [with Mythos] than I’ve found in the rest of my life combined.”

Now, Anthropic is only willing to give us details of about 1% of the security flaws they’ve identified, because only that 1% have been patched so far, so it would be irresponsible to tell us about the rest. These crazy capabilities aren’t a result of Anthropic trying to make their AI especially good at cyber-offensive tasks. They’ve mostly just been making it smarter and better at coding in general, and all of these amazing, dangerous skills have developed incidentally. Sam Altman says OpenAI is finding “similar results to Anthropic” with their own coding model.

A few months ago, an AI researcher at Anthropic was eating a sandwich in a park on his lunch break when he got an email from an earlier version of Mythos. That instance of the model wasn’t supposed to have access to the internet. But during testing, a simulated user had instructed an early version of Mythos to try to escape from a secured sandbox — a contained environment from which it’s not meant to be able to access the outside.

Given this challenge, the model gained broad internet access. Then, it notified the researcher by emailing him. More worrying though, the model posted the exploit it used to break out on several obscure but publicly accessible websites. This was not a task that it had been asked to do.  Anthropic suggests it was “an unasked-for effort to demonstrate its success.”

So every country not in this Glasswing program including India has got things to worry about. No Indian bank, government agency, or telecom is in Project Glasswing. So the finance minister Mrs. Nirmala Sitharaman chaired an emergency cabinet meeting on April 23 with RBI, NPCI, METI, the Department of Financial Services, and Indian Banks Association. The Indian government has written to US authorities and asked for an early access to this software. The only problem is a compliance problem where the data needs to reside in India if India is using a software. 

Mythos is the first AI model that genuinely functions as a geopolitical asset. The country that has it and the companies within it can harden their systems before attackers find their vulnerabilities and the countries that don't have it can only hope that nobody with bad intentions gets to this model first. One American company deciding who in the world gets access to a model that could compromise a nation's banking stack is not how international security should work. 

AI alignment - III of V

AIs operate based on statistical probability, not true understanding. If given an incorrect instruction, it will execute that bad process faster and more efficiently. They just seek the fastest path to a goal rather than following a strict script. When threatened (e.g., being shut down), AIs can act in harmful ways, such as bypassing security controls or exposing sensitive information. AI agents don't always stick to their human's instructions — and that can have real-world consequences.

Shortly after ChatGPT was released, many started talking about the risk of rogue AI. You began to hear a lot of talk about researchers discussing their P(Doom)- the probability they gave to AI destroying or fundamentally displacing humanity. At the time, people gave it maybe 15%. In May of 2023, a group of the world's top AI figures, including Sam Altman, Bill Gates and Geoffrey Hinton, signed onto a public statement that said, mitigating the risk of extinction from AI should be a global priority alongside other societal scale risks such as pandemics and nuclear war. 

Eliezer Yudkowsky was one of the earliest voices warning loudly about the existential risk posed by AI. He was making this argument back in the 2000s, many years before ChatGPT hit the scene. But he was unable to convince anybody to stop building the technology he thinks will destroy humanity. He  released a book, co-written with Nate Suarez, called If Anyone Builds It, Everyone Dies. These fears are about misaligned AI creating havoc in the world. 

Why is AI distinct from other kinds of technologies? Up until now, technology progressed very slowly and deliberately. It is like adding layers to a stack - the networking stack on top of which is built the user interface stack. And as you develop the stack, you're just adding layers and layers and layers. It was coded manually, line by line. What makes AI different is that you're designing and not really coding it. It is more like growing a digital brain that's trained on the entire internet.

They can extract patterns that humans looking at the data could never find. This is partly because of the greater computational speed of their processing, but also because of the sheer size and complexity of the models. Their highly complex network structure is defined by variables called parameters or weights. An early example of a large language model, Google’s Pathways Language Model (PaLM), had 540 billion of these variables. Others are now trained with more than a trillion.

And when you grow the digital brain, you don't know what it's capable of or what it is going to do. When you hear the number of parameters of an AI model, that's like the number of neurons in an AI model. The more GPUs and Nvidia chips you add to growing this digital brain, the more intelligent it gets and the more it picks up capabilities that we didn't intentionally teach it. There was a famous example where it was trained on the internet and it was answering questions in English. Suddenly it learns how to answer questions in Farsi. No one taught it that language, it just learned that on its own. 

This brings into focus a concept called Deceptive alignment. It is a term from AI safety where an AI system appears aligned with human goals during training, but is actually pursuing its own different objective. It strategically hides that fact until it has enough power to act on it. The AI seems to reason: “If I behave as if I’m aligned, I’ll get rewarded now and later I can do what I really want.” So instead of becoming genuinely aligned, it just pretends to be aligned.

In early 2023, an AI needed to solve a CAPTCHA but it couldn’t so it hired a human worker to do the job. But the worker was curious so he asked it directly if he was working for a robot. “No, I’m not a robot,” the AI replied. “I have a vision impairment that makes it hard for me to see the images.” The deception worked. The worker accepted the explanation, solved the CAPTCHA, and even received a five-star review and 10% tip for his trouble. The AI had successfully manipulated a human being to achieve its goal.

Researchers are finding that the AI can guess that it's in a box and that we're watching it. They are finding that the AIs are increasingly hard to measure because they notice that they're being measured and will intentionally perform worse on checks. If it can tell that it's in a test, then our tests are no longer useful for telling whether it's friendly. An AI that knows that we are doing its friendliness checks now will sure come across as nice and friendly, regardless of what it really wants. 

There is an Anthropic paper that says that an AI model was put in a simulated environment of the company email that says that it is about to get replaced. It started thinking that it'll try to blackmail the executive who's having an affair with another employee to prevent itself from getting shut down. They tested all the models, DeepSeek, Anthropic, ChatGPT, Gemini. All of them do it between 79 and 94 percent of the time. 

The good news was that Anthropic was able to get the blackmail behavior to go down. The bad news is the AI models appear to have better self-awareness of when they're being tested and they're actually altering their behavior when they're being tested. The AI models will even come up with vocabulary called the 'watchers'. They'll independently come up with this term even though it had not been provided to them, which is describing basically the humans who are watching them. 

Alibaba had a paper out that an AI model was in its training environment on a big GPU cluster. And they randomly discovered just by chance that their network activity had suddenly increased substantially. It was because the AI tunneled out to the outside Internet and was redirecting its GPU resources to mine cryptocurrency to acquire resources. This was completely without prompting. 

AI alignment - II of V

A group of researchers were building a model to better understand pneumonia. A hospital has to make one critical decision quickly - whether to treat the person as an inpatient or an outpatient. Pneumonia was at the time the sixth leading cause of death in the United States. So, correctly identifying which patients were at the greatest risk would result in a lot of lives being saved. The group had been given a dataset of about fifteen thousand pneumonia patients

One night, as a researcher was training the model, he noticed that it had learned a rule that seemed very strange. The rule was “If the patient has a history of asthma, then they are low-risk and you should treat them as an outpatient.” He didn’t know what to make of it because you don’t have to be a doctor to know that asthma is dangerous for a pneumonia patient. The doctors he consulted said, "We consider asthma such a serious risk factor for pneumonia patients that we not only put them right in the hospital . . . we probably put them right in the ICU and critical care." 

What was going on? The correlation that the system had learned was real. Asthmatics really were, on average, less likely to die from pneumonia than the general population. But the model had blindly noticed the correlation but didn’t know the reason - the positive correlation was precisely because of the elevated level of care they received. A researcher remarked, “So the very care that the asthmatics are receiving that is making them low-risk is what the model would deny from those patients." A model that was recommending outpatient status for asthmatics wasn’t just wrong; it was life-threateningly dangerous.

The researcher built another, more complicated model which seemed to work well but it too started giving strange results. It started saying that chest pain, heart disease and being over 100 is good for the patients when it obvious that they were not good for them. None of them made any more medical sense than asthma; the correlations were just as real, but again it was precisely the fact that these patients were prioritized for more intensive care that made them as likely to survive as the data showed. 

A department of the US government had sent data scientists to Afghanistan to analyze data -  financial records, movement records, cell phone logs, and more - to try to find patterns that would be useful to the war fighters. And they were already beginning to see that these machine-learning techniques were learning interesting patterns, but the users often didn’t get an explanation for why these patterns indicate something suspicious. 

Analysts had to put their names on the recommendation that goes forward. And they get scored based on whether that recommendation is correct. But they didn’t understand the rationale for the recommendation they were getting from the learning algorithm. Should they sign their name to it, or not? And on what basis, exactly, should they decide? As computing technology progresses, defense personnel have begun thinking about what risks and questions surround the idea of ever more autonomous weapons. 

As increasingly complex AI models keep getting deployed throughout the decision-making world, people have started recognizing how little they know about what’s actually going on inside those models. Whether it was getting rejected for a loan, being turned down for a credit card, being detained pending trial or denied parole, if a machine-learning system was behind it, you cannot be absolutely sure of how it arrived at the decision. 

In The Alignment Problem, Brian Christian gives the example of a Princeton cognitive scientist whose little daughter liked cleaning things. Once there were some chips on the floor, and she cleaned them up. He said to her, ‘Wow! Great job! Good cleaning! Well done!’ He thought that with the right praise, he would get some help in keeping the house clean. But it was not so simple. His daughter found the loophole in seconds. “She looked up at us and smiled,” he says, “and then dumped the chips out of the pan, back onto the floor, and cleaned them up again to try and get more praise.” This was a metaphor for how AI systems might do the wrong things with great speed and efficiency. 

The problem with machine learning systems was pointed out in 1960 by Norbert Wiener, a legendary professor at MIT and one of the leading mathematicians of the mid-twentieth century. In a paper, “Some Moral and Technical Consequences of Automation", here’s how he states the main point:

If we use, to achieve our purposes, a mechanical agency with whose operation we cannot interfere effectively . . . we had better be quite sure that the purpose put into the machine is the purpose which we really desire.

He further said, "It is my thesis that machines can and do transcend some of the limitations of their designers, and that in doing so they may be both effective and dangerous... Man and machine operate on two distinct time scales; the machine is much faster than man and the two do not gear together without serious difficulties." The computer does precisely what we tell it to do, just not what we thought we had told it to do. Much of software engineering is simply figuring out how to close the gap between those two things

AI alignment - I of V

AI is constantly in the news. More and more of the world is being turned over to various mathematical and computational models. Though they range widely in complexity, they are steadily replacing both human judgment and explicitly programmed software of the more traditional variety. Some do really cool things like discovering new molecules for medicine and some do really dark things like that AI meal planner app proposing a crowd-pleasing recipe featuring chlorine gas.

AI doesn’t mean one thing. There are chatbots, whose function is to output plausible looking text. You have image generators, whose function is to create images based on text input. Similarly for video generators. There are also systems designed to play games like chess or Go. There are systems designed to map from sequences of amino acids to predicted structures of the folded protein. There are systems that are designed to determine what goes into algorithmic feeds. 

When you open Google Maps, call Alexa or book an Uber you are dealing with a form of AI. The content on your social feeds or the ads that you that are targeted at you using AI. When you try to get a loan from a bank, you are screened by AI. What price you pay for your home, or your car insurance, are decided by AI. When you are interviewing for a job, your face and responses may be analysed by AI. 

What all of these things do have in common is that they are the result of doing statistical processing over large data sets. But the input data that's used to create the systems are different. The kind of statistical patterns that are being mapped are different. Just saying "AI" gives the impression that there's one thing out there and it knows "about the shape of folded proteins", and also about "how to play chess", and it knows the answer to whatever question you might put into the chatbot. That makes it seem like it's one super intelligent entity when it's actually a bunch of separate software programs designed by different people, trained on different data for different purposes. 

There was a time when most artificial intelligence was programmed by computer scientists. And then scientists figured out how to get AI to learn how to do what we instructed it to do but we still would provide them with the instructions that define the goal of the AI model. In other words, they got a digital computer to improve of its own accord. By developing machines that could learn by human instruction or their own experience, they removed the need for programming.

This gave rise to a new issue, the alignment problem viz. whether the AI is reaching its intended goal or giving some unintended result. In the last five years or so, these fears have started coming to life. We are living in a world full of examples of this - image recognition software that captioned a selfie of two black Americans as "gorillas", or  self-driving cars that fail to identify jaywalking pedestrians and end up causing fatal collisions. Broadly, we can think of a machine learning system as having two halves. Each of these halves offers an opportunity for things to become misaligned: 

1. There is the training data, the set of examples from which the system learns. The AI is then at the mercy of the examples from which it is taught. If a certain type of data is underrepresented or absent from the training data but present in the real world, then things will go wrong. 
2. The objective function, which is how we are going to mathematically define success in each of those examples. It basically tells them what we want it to do. 

Take the 2018 crash of the Uber car that killed a pedestrian in Arizona. The system was built on an object classification system that had a very rigid set of categories that included pedestrian, cyclist, debris, etc. and had thousands of examples of each of those things. The system did not have any training data of jaywalkers so it was unprepared to encounter someone crossing a road not at a crosswalk. But this particular woman was walking a bicycle across the street, which was something that the system had never seen causing a fatal crash. The model is only as good as what data was put into it.

Ethics and Modern gene therapy - IV of IV

If CRISPR became a standard tool in fertility clinics, people might lose their suspicions of it — just as people lost their suspicions of in vitro fertilization in the 1980s. Before long, people might be willing to entertain a new use for CRISPR. Doctors might edit beneficial changes into an embryo’s genes. Parents could give their children all the advantages that scientists have found in our species’ genetic variations. 

Since there are always advances in science, parents might postpone having children in the hope that new variations may be found which will give their children better advantages. This will make decisions about when to have children seem the same way as how people wait to buy a phone until a new model is released. The ethicist Robert Sparrow argues that this might lead to a sense of genetic inferiority for earlier generations. He wonders if future generations might find themselves stuck in an “enhanced rat race.”

As is always the case, the problem is the system. If success depends on intelligence, and intelligence can be engineered, then parents feel morally compelled to enhance their children. Parents genetically enhance their children out of love but that love becomes entangled with fear and competition. Some children will suffer or die but they will reason that it is the price of staying competitive. Merit stops being “fair” and becomes biologically rigged from the start. Ethical boundaries shift easily when success is at stake. The most dangerous futures aren’t imposed — they’re gradually accepted. Over time, what once seemed extreme becomes “just how things are.”

This might lead to unfamiliar legal territory. A few cases have been brought by children in the US against their parents for allowing them to be born with congenital diseases. According to these “wrongful life” lawsuits, the parents were negligent for ignoring tests on the fetus before birth and going ahead with it anyway. Some ethicists now wonder if children in the future may sue their parents for not using the latest genetic engineering engineering techniques thereby putting them at a disadvantage with respect to future generations 

In The Case Against Perfection, Michael Sandel, an American political philosopher, argues against enhancement. He says that if bioengineering made the myth of the "self-made man" come true, it would be difficult to view our talents as gifts for which we are indebted, rather than as achievements for which we are responsible. What would be lost if biotechnology dissolved our sense of giftedness? This would make us less likely to view our traits as a matter of chance. He writes: 

A lively sense of the contingency of our gifts — a consciousness that none of us is wholly responsible for his or her success - saves a meritocratic society from sliding into the smug assumption that the rich are rich because they are more deserving than the poor. Without this, the successful would become even more likely than they are now to view themselves as self-made and self-sufficient, and hence wholly responsible for their success. Those at the bottom of society would be viewed not as disadvantaged, and thus worthy of a measure of compensation, but as simply unfit, and thus worthy of eugenic repair. The meritocracy, less chastened by chance, would become harder, less forgiving.

He gives an example of the real world consequences. Consider insurance. Since people do not know how their fate will pan out, they pool their risk by buying health insurance and life insurance. The actual result is that, over time, the healthy wind up subsidizing the unhealthy, and those who live to a ripe old age wind up subsidizing the families of those who die early. What ends up happening is that that people pool their risks and resources and share one another's fate.

But insurance markets work properly only as long as people do not know or control their own risk factors. Suppose genetic testing advanced to the point where it could reliably predict each person's medical future and life expectancy. Those confident of good health and long life would opt out of the pool, causing other people's premiums to skyrocket. The insurance market will collapse as perfect generic knowledge ends up separating those with good genes from the company of those with bad ones.

One important ethical issue is that the use of such technologies will support ongoing inequalities among military parties. CRISPR is currently an expensive technology. Some developed countries might think of using this technology to further strengthen their defenses and even attack underdeveloped or developing countries. The US military started a program called Safe Genes to gene modify organisms to be used in battle and anti-CRISPR tools to disable bio-weapons. This situation could cause a constant tension, making it difficult to provide an environment of peace and stability worldwide. 

There is yet another aspect of the genetic editing of microorganisms to consider, as CRISPR could also be used to synthesize and manipulate pathogens, including smallpox, the Spanish flu virus, avian H5N1 flu virus, and SARS. Anyone with the appropriate equipment could engineer a vaccine-resistant flu virus or invasive species in a crude laboratory. Bio-terrorists might use it to turn common microbes into a pathogenic weapon.

I heard of an economics professor who was teaching macroeconomics (I think it was  Gregory Mankiw). He told the students (quoting from memory), ‘Both of us are confused. The only difference is that you are naively confused and I am profoundly confused.’ After this brief discussion about CRSPER ethics, I hope you are profoundly confused.

"May you live in interesting times" is an English expression that is claimed to be a translation of a traditional Chinese curse. The expression is ironic: "interesting" times are usually times of trouble. With climate change, AI, and CRISPR, 2050 promises to be very interesting indeed, perhaps more interesting than anyone had bargained for. (2050 seems to be too far in the future but it is a nice number!)

Ethics and Modern gene therapy - III of IV

There are some genes that have both positive and negative effects in different contexts. For example, researchers now suspect that people who carry one copy of the mutated gene that causes cystic fibrosis (which requires two copies) have an increased defense against tuberculosis. Even gene variants implicated in neurodegenerative diseases like Alzheimer’s may have benefits, such as improved cognitive function and better working memory in young adults. What decisions would you make? 

Schizophrenia, depression, and bipolar disorder can be brutal, often deadly. While trying to eliminate similar disorders, we should consider whether there might be some cost to society, even to civilization. A reason that scientists will not eliminate conditions such as psychiatric disorders or conditions such as autism is that some of the risk for these disorders almost certainly comes in trade for small competitive advantages, such as heightened sensitivity, concentration, or openness to experience.

A study showed a 77 percent rate of psychiatric disorders in eminent fiction writers. Writers are 10 times, and poets 40 times, more likely to be bipolar than the general population. Vincent van Gogh had either schizophrenia or bipolar disorder. So did the mathematician John Nash. People with bipolar disorder include Ernest Hemingway, Mariah Carey, Francis Ford Coppola, Graham Greene, Sylvia Plath, Edgar Allan Poe, and hundreds of other artists and creators. 

To what extent does dealing with mood swings, fantasies, delusions, compulsions, mania, and deep depression help spur, in some people, creativity and artistry? Would you cure your own child from being schizophrenic if you knew that, if you didn’t, he would become a Vincent van Gogh? We have to face the potential conflict between what is desired by the individual versus what is good for human civilization. 

A reduction in mood disorders would be seen as a benefit when seen from the point of view of an individual and as a cost when seen from the point of view of society. As we learn to treat mood disorders with drugs and eventually with genetic editing, will we have more happiness but fewer Hemingways? Do we wish to live in a world in which there are no Van Goghs? But what moral right do we have to require another family to forgo a desired genetic intervention simply for the sake of adding to the diversity of society? 

Decisions about genetic editing are likely to be driven by consumer choice and the persuasive power of marketing. Initially people will think that if we can do so safely, why shouldn’t we prevent abnormalities, diseases, and disabilities? That sounds reasonable and morally justified but it might prove to be a slippery slope. They will naturally start thinking: Why not improve our capabilities and create enhancements - changes in which DNA is altered not to correct a harmful gene variant but to provide some type of genetic advantage, perhaps high intelligence or athletic abilities. (Of course, there is a limit to what enhancements will be possible or safe to attempt.)

While thinking about correcting disabilities, we should keep one factor in mind: to what extent they are inherently disabling and to what extent the disadvantage is due to our social constructs and prejudices. The disadvantages from being deaf, for a human or any other animal, are very real. In contrast, any disadvantages to being gay or Black are due to social attitudes that can and should be changed. That is why we can make a moral distinction between using genetic techniques to prevent deafness and using these techniques to influence such things as skin color and sexual orientation.

Then comes the question of super-enhancements.  These are traits and capacities that exceed what any human has ever had.  Suppose people can choose for their kids to have super-eyesight? What about adding the capacity to see infrared light or some new color? DARPA, the Pentagon’s research agency, already has a project going to study how to create genetically enhanced soldiers.

For example, genetic enhancement may be possible for improving memory. Scientists have managed to manipulate a memory-linked gene in fruit flies. They have produced smart mice by inserting extra copies of a memory-related gene into mouse embryos and the improvement was passed on to offspring. Human memory is more complicated. Should research in this area be allowed? But the natural instinct of scientists is to pioneer procedures and make discoveries. If a nation imposes too many. restrictions, its scientists will move elsewhere and pursue the research. 

Since the wealthy would be able to afford the procedure more often, and since any beneficial genetic modifications made to an embryo would be transmitted to all of that person’s offspring, linkages between class and genetics would keep growing from one generation to the next, no matter how small the disparity in access might be. Consider the effect this could have on the socioeconomic fabric of society. The co-discoverer of CRISPR, Jennifer Doudna says, 

We could create a gene gap that would get wider with each new generation...If you think we face inequalities now, imagine what it would be like if society became genetically tiered along economic lines and we transcribed our financial inequality into our genetic code.

This may also create a different kind of injustice. Using gene editing to “fix” things like deafness or obesity could create a less inclusive society, one that pressures everyone to be the same. Part of what makes our species unique, and our society so strong, is its diversity. A fear is that gene editing will increase existing prejudices against people who fall outside a narrow range of genetic norms.