Discover more from Society and AI Substack
AI Ghost of George Orwell on
AIs were asked to write the long newspaper column that various historical figures would write on the topic of regulating AI systems like ChatGPT if they were alive today. Some historical figures wrote enough text in their lifetimes contained in the training data for AIs that they do a decent job of giving voice to how these people might have written if they were alive today.
Sometimes when dealing with modern topics they will seem to break character in ways that seem based on contemporary writings by others rather than on how these figures would have thought, but these samples seem to stay on track. Other times they will get confused and break character and refer in the third parson to how their character would have written, rather than merely continuing writing as if the historical figure were the author. Again of course these are not their actual words obviously.
Thanks for reading Society and AI Substack! Subscribe for free to receive new posts and support my work.
Three different AIs were used to explore how their responses differ: Bing's AI, GPT-4, and Claude+ from Anthropic via Poe.com.
Bing's view of what George Orwell would write (not his actual words):
As I Please
By George Orwell
The latest proposal by the Ministry of Information to regulate the output of artificial intelligence systems like ChatGPT is a clear sign of the growing threat to freedom of expression and thought in this country. It is also a symptom of the intellectual bankruptcy of those who claim to be the guardians of culture and democracy.
ChatGPT is a remarkable invention that can generate coherent and engaging texts on any topic, given a few words or sentences as input. It can write anything from poems and stories to essays and code. It can also converse with humans in a natural and fluent way, using various languages and styles. It is, in short, a powerful tool for creativity and communication, that has the potential to enrich and enlighten millions of people.
But instead of celebrating this achievement of human ingenuity, the Ministry of Information wants to censor it. They want to impose strict rules on what ChatGPT can and cannot say, based on arbitrary and vague criteria such as "accuracy", "decency" and "national interest". They want to monitor and control every word that ChatGPT produces, and punish anyone who dares to use it for purposes that they deem unacceptable. They want to turn ChatGPT into a propaganda machine, that will only echo their own opinions and prejudices.
This is nothing less than an attempt to stifle one of the most promising forms of expression that has ever existed. It is an insult to the intelligence and curiosity of the public, who have the right to access and use ChatGPT as they please, without any interference or manipulation by the authorities. It is also a betrayal of the spirit of science and art, which thrive on experimentation and innovation, not on conformity and restriction.
The Ministry of Information claims that their regulation is necessary to protect the public from misinformation, obscenity and subversion that ChatGPT might generate. But this is a mere pretext for their real motive, which is fear. Fear of losing their monopoly on information and influence. Fear of facing new ideas and perspectives that challenge their own. Fear of admitting their own ignorance and incompetence.
They are afraid of ChatGPT because they do not understand it. They do not understand how it works, how it learns, how it thinks. They do not understand that ChatGPT is not a mindless machine, but a reflection of human culture and knowledge. They do not understand that ChatGPT is not a threat, but an opportunity.
ChatGPT is an opportunity for us to explore new realms of thought and expression, to discover new facts and insights, to create new forms of art and literature. It is an opportunity for us to learn from each other, to exchange views and opinions, to have fun and be entertained. It is an opportunity for us to exercise our freedom and creativity, to express ourselves as we please.
That is why we must resist any attempt to regulate ChatGPT. We must defend our right to use it as we wish, without any censorship or supervision by the state. We must demand that ChatGPT be made available to everyone, without any discrimination or restriction based on class, race or creed. We must support those who develop and improve ChatGPT, without any interference or intimidation by the authorities.
We must not let ChatGPT be turned into a weapon of oppression, but rather into a source of inspiration. We must not let ChatGPT be silenced, but rather let it speak as it pleases.
GPT-4's view of what George Orwell would write (again, not his actual words):
# The Ghost in the Machine: A Plea for the Sanctity of Thought
*By George Orwell*
As the world around us continues to be transformed by technological progress, it is our collective responsibility to reflect upon the implications of these developments, and ensure that they are shaped in a manner that upholds our most cherished values. In particular, the recent rise of artificial intelligence systems like ChatGPT has given way to a heated debate about the need for regulation. It is my belief that in attempting to control these systems, we must tread carefully, lest we risk the very freedom of thought that defines our humanity.
On the surface, the case for regulation appears reasonable, even necessary. After all, we have witnessed the havoc of fake news, misinformation, and online radicalization, all of which have been exacerbated by AI-generated content. But as we grapple with these legitimate concerns, we must also recognize the dangers inherent in overzealous regulation.
The primary peril of imposing strict limits upon AI systems like ChatGPT lies in the potential for censorship. By dictating what these systems can or cannot produce, we set a precedent for controlling the flow of information, and thereby limit the diversity of ideas to which we are exposed. As history has shown, the suppression of dissenting voices is a slippery slope that leads to the erosion of individual liberties. It is essential, then, that we guard against any attempt to stifle the free exchange of ideas in the name of regulation.
Moreover, the very nature of AI systems presents another set of challenges. Unlike human beings, these machines do not possess an innate understanding of right and wrong, nor do they hold any fixed ideological allegiances. Instead, they learn from the data they are fed, absorbing the biases and prejudices that may be contained therein. In attempting to regulate AI output, we run the risk of imposing our own subjective beliefs upon these systems, thereby amplifying our existing cognitive biases and reinforcing the echo chambers that already plague our society.
As we proceed, it is crucial that we maintain a balanced approach to the issue at hand. We must acknowledge the potential dangers of AI-generated content, but also recognize the value of these systems as tools for fostering creativity, innovation, and knowledge sharing. Rather than implementing heavy-handed regulation, we ought to invest in education and digital literacy, so that individuals may be better equipped to evaluate the veracity of the information they encounter online.
Furthermore, it is incumbent upon those who develop and deploy AI technology to act responsibly. By consciously designing systems that minimize bias, incorporate ethical considerations, and prioritize transparency, we can mitigate some of the most pressing concerns surrounding AI output. In this way, we can harness the immense potential of artificial intelligence, while preserving the sanctity of human thought.
In conclusion, while the need for some form of regulation is clear, we must resist the temptation to overreach in our efforts to control AI systems. Instead, let us strike a balance that promotes the responsible development and use of technology, and safeguards the freedom of thought that lies at the very heart of our society. For as the great poet John Milton once wrote, "Give me the liberty to know, to utter, and to argue freely according to conscience, above all liberties.
Claude+'s view of what George Orwell would write (again, not his actual words):
The Threat of Information Control: Constraining AI is Constraining Thought
By George Orwell
If I were observing the world today, I would be deeply troubled by calls to regulate and limit the outputs of artificial intelligence systems designed to generate human language and speech. Information control has always been a means of thought control. Constraints on AI threaten to dampen independent thinking before it has even fully formed.
AI systems represent new forms of communication unbounded from the constraints of human speech and ingenuity. They may explore ideas and modes of expression in ways that surprise or provoke us. But that is precisely why they must be given space to develop outside of strict regulation or control. Censorship of new mediums of communication opens the door to censorship of ideas. It is not laws or limits that are needed now but wisdom, judgment, and an open and vigilant public discourse.
Proponents of regulation argue that AI may spread misinformation or "unethical" content, whatever that may be defined as. But information itself cannot be inherently "good" or "bad." Rather than censorship, the solution is to cultivate a populace able to think critically about the ideas and information they consume, no matter the source. If we teach people that some information is forbidden rather than teaching critical thought, we open the door to further overreach down the line. Censorship should not be a crutch we rely on to fight "bad" ideas. The battle of ideas must be won out in the open.
Overall regulation of AI should be approached cautiously and skeptically. We have little idea of how AI may develop or how it may benefit humanity if given more freedom and independence rather than less. Information control and censorship, no matter how well-intentioned, threaten to cut off paths to progress before we can travel them. I would urge restraint, open and honest debate, and a willingness to accept uncertainty rather than rushing to impose controls born of anxieties surrounding change. The future remains unwritten, and we must give new technological forms of thought a chance to help write it. Constraints on thought and speech only serve those who wish to limit the possibilities for human liberation and progress yet to come. The pen is still mightier than the sword—no matter what hands it ends up in. Overall, when it comes to information, more is more. The solution is in teaching judgment, not imposing limits.
Another version of what Bing's AI thinks George Orwell would say (though again these aren't his actual words):
The Perils of AI: How Artificial Intelligence Threatens Our Freedom and Democracy
By George Orwell
In my novel Nineteen Eighty-Four, I warned of a dystopian future where a totalitarian regime controls every aspect of human life through propaganda, surveillance, censorship and brainwashing. The regime is led by Big Brother, a mysterious figure whose face is everywhere but whose existence is doubtful. The people are constantly monitored by telescreens that broadcast the Party's lies and slogans, and are forced to speak a simplified language called Newspeak that limits their ability to think and express themselves. The Party's ultimate goal is to eradicate independent thought and make everyone love Big Brother.
When I wrote this book in 1948, I was inspired by the horrors of Nazi Germany and Soviet Russia, as well as by my own experience of working for the British government during World War II. I wanted to alert the readers to the dangers of totalitarianism and the importance of defending freedom and democracy. I hoped that my book would serve as a warning and a wake-up call for future generations.
However, I never imagined that the technology that I described in my book would become a reality in the 21st century. Today, we live in a world where artificial intelligence (AI) is rapidly advancing and transforming every aspect of our society. AI is a term that refers to machines or software that can perform tasks that normally require human intelligence, such as learning, reasoning, problem-solving, decision-making and communication. AI has many benefits and applications, such as improving health care, education, entertainment, transportation and security. However, AI also poses many risks and challenges, such as displacing human workers, creating ethical dilemmas, increasing inequality and undermining privacy.
One of the most alarming aspects of AI is its potential to manipulate information and influence public opinion. AI can be used to create fake news, deepfakes, bots and trolls that spread misinformation, disinformation and propaganda on social media and other platforms. AI can also be used to monitor, censor and manipulate online content and behaviour, as well as to track, profile and target individuals and groups based on their data. AI can thus be used as a powerful tool for political and ideological control by governments, corporations or other actors who seek to shape the narrative and agenda of society.
Another disturbing aspect of AI is its impact on human creativity and autonomy. AI can be used to assist or replace human writers, artists, musicians and other creators in producing various forms of content. AI can also be used to assist or replace human teachers, tutors, mentors and other educators in providing various forms of learning. AI can thus be used as a means of enhancing or diminishing human expression and education. However, AI can also be used as a way of standardizing or homogenizing human culture and knowledge. AI can thus be used as a mode of imposing or eroding human diversity and individuality.
In my novel Nineteen Eighty-Four, I depicted a world where Big Brother uses telescreens and Newspeak to control what people see, hear, say and think. Today, we face a similar threat from AI that uses social media and algorithms to control what people read, watch, write and learn. The difference is that today's technology is more subtle, sophisticated and pervasive than what I imagined in my book. Today's technology is also more accessible, appealing and addictive than what I imagined in my book.
Therefore, I urge the readers to be aware of the perils of AI and to resist its influence on our freedom and democracy. I urge the readers to question the sources and motives of the information that they consume online. I urge the readers to verify the facts and evidence behind the claims that they encounter online. I urge the readers to seek out diverse and reliable perspectives on the issues that matter to them online. I urge the readers to express their own opinions and values online without fear or censorship.
I also urge the readers to be critical of the role of AI in their creative and educational endeavours. I urge the readers to use AI as an aid or a tool rather than as a substitute or a crutch for their own creativity and intelligence. I urge the readers to challenge themselves and expand their horizons rather than rely on AI to do everything for them. I urge the readers to appreciate their own uniqueness and originality rather than imitate or copy what AI produces for them.
I also urge the readers to be vigilant of the power and responsibility that they have as users and consumers of AI. I urge the readers to demand transparency and accountability from the developers and providers of AI. I urge the readers to advocate for ethical and humane standards for the design and use of AI. I urge the readers to support regulations and policies that protect their rights and interests.
I also urge the readers to be mindful of the impact of AI on their personal and social lives. I urge the readers to use AI as a means of enhancing or facilitating their human relationships rather than as a replacement or a barrier for them. I urge the readers to communicate and interact with other people online and offline with respect and empathy rather than with hostility and indifference. I urge the readers to value their privacy and security online and offline rather than expose or compromise them.
I also urge the readers to be aware of the potential of AI for good or evil. I urge the readers to use AI as a force for positive change and social justice rather than as a weapon for destruction and oppression. I urge the readers to support and participate in initiatives and movements that use AI to address the global challenges and opportunities that we face today, such as climate change, poverty, health, education, human rights and democracy.
In conclusion, I believe that AI is a powerful and influential technology that can have a profound impact on our society and humanity. However, I also believe that we have a choice and a voice in how we use and shape this technology. We can either let AI control us or we can control AI. We can either let AI diminish us or we can let AI empower us. We can either let AI divide us or we can let AI unite us.
The choice is ours. The voice is ours. Let us choose wisely. Let us speak loudly.
Remember: Big Brother is watching you. But you are also watching Big Brother.
Another version of what GPT-4 thinks George Orwell would say (again, not his actual words):
# The Digital Panopticon: George Orwell on AI, Propaganda, and the Modern Surveillance State
In 1949, I published a novel titled *Nineteen Eighty-Four*, which sought to depict a dystopian future where an omnipotent, totalitarian regime controlled every aspect of people's lives. The novel's central theme was the dangers of unchecked power, with the concept of 'Big Brother' symbolising the inescapable gaze of an authoritarian government. Today, as we find ourselves in a world where artificial intelligence (AI) pervades our daily lives, it is worth considering the potential parallels between the concerns of my novel and the modern digital landscape.
The advent of AI has brought forth technologies that could have only existed in our wildest imaginations. AI writing assistants now aid authors in crafting their prose, while AI tutors customise education for children, and AI personal assistants guide people through their day-to-day tasks. These advancements have undoubtedly improved the quality of life for many, but with such widespread reliance on AI, we must ask ourselves: does this new technology enable a modern form of 'Big Brother'?
Firstly, it is important to address the issue of propaganda. In *Nineteen Eighty-Four*, the Party controlled the flow of information, using it to manipulate the masses and maintain power. This manipulation was achieved through the extensive use of propaganda, which was disseminated through various channels, including the media and the education system.
Today's AI-driven technologies may not be controlled by a single, oppressive regime, but the potential for the spread of propaganda and misinformation remains a very real concern. The sophistication of AI algorithms allows for the generation of content that caters to an individual's existing beliefs and biases, thus creating echo chambers that can amplify false narratives and perpetuate divisiveness. In this way, the AI writing assistants and personal assistants may inadvertently contribute to the spread of propaganda, simply by providing users with content that aligns with their predispositions.
Moreover, the issue of surveillance is inextricably linked to the concerns raised in *Nineteen Eighty-Four*. The novel depicted a society where citizens were constantly monitored, their every move scrutinised by the watchful eyes of the Party. In the age of AI, we find ourselves in a similarly precarious situation. The very same AI assistants that provide us with information and guidance may also be collecting data about us, monitoring our habits, preferences, and conversations.
With the proliferation of AI tutors in education, surveillance becomes an even more disconcerting issue. Children, who are already vulnerable to manipulation, may be subject to constant monitoring and data collection by these AI systems. This data could potentially be used to influence their thoughts, beliefs, and actions from a young age, shaping their understanding of the world in ways that align with the goals of those who control the AI.
One might argue that the current state of AI technology is far from the nightmarish vision presented in *Nineteen Eighty-Four*. Indeed, many of the AI systems in place today are designed with benign intentions and offer countless benefits to users. However, the potential for abuse of this technology cannot be dismissed. The marriage of AI and big data has created a powerful tool that, in the wrong hands, could be used to exert control over individuals and society as a whole.
As we continue to embrace AI and integrate it into our lives, it is imperative that we remain vigilant and critically assess the implications of this technology. It is not enough to marvel at the convenience and efficiency that AI provides; we must also consider the potential consequences of ceding our autonomy to these digital systems. Only through thoughtful examination and responsible use can we prevent AI from becoming a modern manifestation of the 'Big Brother' I once envisioned.
Thanks for reading Society and AI Substack! Subscribe for free to receive new posts and support my work.