AI is everywhere. Shop online, use a bank card in a store, Google some information, watch a Netflix movie, visit Instagram, Facebook, Spotify or TikTok, speak to Alexa or try out ChatGPT – AI is there behind the scenes, assessing, analyzing, monitoring and predicting your behaviour.
Get into your car, and AI-powered number plate recognition springs into action, checking out databases and informing the police or the DVLA if you are wanted for questioning. Walk into many retail outlets and covert AI-powered facial recognition will identify if you are a known shoplifter. If you apply for a job with a large company, AI will assess your CV and decide whether you should be short-listed. Apply for a bank loan or a mortgage, and AI will evaluate your creditworthiness. If you have a medical scan, AI will assist the health professionals in deciding whether there’s a problem.
But what exactly is it?
The term artificial intelligence AI has been around for decades – tracing back to a key US workshop in 1956 when a number of leading computer experts first started using the terminology for their research. Their goal was to use computer technology to simulate and reproduce human learning and intelligence.
For many years research in university laboratories AI made painfully slow progress, but just in the last 10 years the field has exploded, thanks to software developments in machine learning – computer programs based on a simplified representation of a network of brain cells.
Each program is trained on large quantities of data and is capable of a highly sophisticated form of pattern recognition – think face recognition, for example. “Clearview” is one face recognition program which has a database of more than 20 billion images collected from the internet, including social media. The program is currently being used by US law enforcement and government agencies to identify individuals using surveillance cameras, smartphones and social media apps. It is obvious that this powerful technology raises complex ethical and social issues and concerns about privacy and abuse of the technology have led to the Clearview program being banned in several other countries.
Just in the last 12 months the tech world has been abuzz because of explosive developments in a new form of software called ‘generative AI’. These are mind-bogglingly complex programs that can create original text, audio, images and video, on the basis of patterns learned from vast amounts of training data.
ChatGPT, a text generating program, was made available to the general public in December 2022 and it is said to have been the fastest-growing internet service ever, reaching 100 million users two months after launch. OpenAI, the company that created the program, struck a $10 billion deal with Microsoft, and this technology is now built into Office software and the Bing search engine. Google, Microsoft, Amazon and other tech companies are in a desperate race to create generative programs which are ever more sophisticated and human-like. We are moving into a world where we may never know whether an article, a sermon, a student essay, a scientific paper, even a book, was created by a human being, a mindless AI system or some combination of the two.
And AIs are not only becoming scarily good at creating original text but also images, videos, document design, architectural plans, music, computer code, you name it.
Progress or problematic?
The implications for educators, academics, authors, journalists and creative artists are impossible to predict. Some computer specialists are calling this new technology as significant as the internet itself. But there’s no doubt that there are serious problems and dangers looming ahead. If AI can write an essay or plan a menu for you, could it be used to create powerful ‘prophetic’ sermons or even new interpretations of biblical truth?
Christian leaders and teachers who are tempted to use these extraordinary tools need to be particularly cautious. ChatGPT and similar programs (known as large language models) are capable of spouting out information in a supremely confident and fluent manner. But some of it has turned out to be false or highly misleading.
The programs are well-known to generate falsehoods, conspiracy theories and heresies. And it is not just factual inaccuracies. The programs are capable of subtle and covert manipulation of ideas and content. The underlying algorithms have been trained on vast amounts of text and other data strewn across the internet, much of it containing hidden biases, deliberate misinformation and disinformation, and also hideous and overt evil. At the same time commercial companies are racing to find ways to embed subtle advertising content into the text, in order to maximise their own profits.
And the precise mechanisms by which ChatGPT selects particular words in response to a specific prompt are completely opaque. Even the engineers who designed and wrote the code behind the program could not tell you why those individual words were chosen. It is all lost within the ultimate ‘black box’ of mindless algorithmic mathematics. How confident can you possibly be about the origin and nature of the text that you are being invited to incorporate as your own words?
Authenticity matters
So how should Christians respond to these powerful tools?
The debate is just beginning and we at the AI Christian Partnership hope to provide thoughtful resources which will help Christian leaders navigate this complex and rapidly changing area. What does safe and appropriate use of AI look like? What guard-rails should be established? How can the most vulnerable in our communities be protected from new forms of abuse and manipulation?
It’s surely important that we avoid adopting fascinating new technologies without thought and spiritual discernment. From a Christian perspective, however attractive AI-generated text may be, we know that authenticity, personal integrity, honesty and truthfulness all matter. This surely applies especially to what Christian leaders write and say in public. So perhaps those of us who want to defend the Christian values of authenticity and truthfulness will have to take a solemn vow to be transparent in our own use of AI-generated material.
Additionally, as AI technology continues to develop and penetrate our lives at breakneck speed it raises two fundamental questions:
1. What does it mean to be human in a world that is increasingly dominated by nearly human machines?
2. How can we direct this powerful technology in line with God’s purposes and towards human good?
These are the questions we will return to, as advancing AI technology transforms our world, our workplaces, our Christian communities and our daily lives.