By Rev. Dr. Timothy Ehrlich

This March 2024 has been extremely eventful for the world of AI and robotics. Calling 2024 the year of the ‘humanoid’ robot, the CEO of Nvidia (the world’s largest AI technology company) Jensen Huang took the stage in their developer conference with 9 humanoid robots from different manufacturers including Boston Dynamics, Figure AI, Fourier Intelligence, Sanctuary AI, Unitree Robotics and XPENG Robotics. All of them are using Nvidia microchips and processors to be the brains of their humanoid robots. Tesla also displayed their new humanoid robot Optimus 2 in March as did two other Chinese companies. They are all rushing to display their products because it is projected that the market for humanoid AI robots will be about one-billion units by 2040![1]         

Also in March, Nvidia announced that they are partnering with Hippocratic AI to create AI nurses that provide video/tele health calls which significantly outperform human nurses in identifying correct medicines and dosages and only cost $9 per hour to operate.[2] 

Neuralink, a subdivision of Tesla revealed that a robotic surgeon had implanted 1000 electrodes in the brain of quadriplegic Noland Arbaugh, enabling him to control the cursor on a computer screen with his mind. Elon Musk is hoping that in the future Neuralink will be commonly used to enhance people’s ability to think; “a sort of merger of biological intelligence and machine intelligence.”[3]  

Musk also believes that AI empowered robots will usher in a utopian “age of abundance” future when every job will be done by robots, and that there will be a universal high income, allowing people to only work at things they want to.[4] Amazon CEO Jeff Bezos agrees, he said of AI, “It is a renaissance, it is a golden age.”[5]

Since 1965, the number of transistors in a device has doubled every two years. This held true for 57 years until 2022. Since 2022 chip processing capacity has been increasing tenfold every year! A good transistor radio has 8 transistors, the newest Nvidia chip has 200 billion transistors and can process 20,000 trillion instructions or calculations per second!

This chart, released by NVIDIA in their March 2024 developers conference shows the dramatic rise in the performance of computer chips since 2022 a TFLOP (teraflop) means the ability to process one trillion operations per second.

The rapid advancements in AI are being driven by four new technological developments that, joined together, have formed super powerful computers that are able to equip humanoid robots with the ability to be both excellent factory workers or domestic servants and companions. How great would it be to have a robot for the home that can cook and clean and care for the aged and infirm or tutor children in academics and help them develop their curve ball or free throw?!

 There are four new technological developments that have made general purpose humanoid robots possible:

1          Super high memory chips allow the machines to store enough information to power chat bots to be able to have an interactive conversation in the way humans do but with much greater speed, accuracy and content of knowledge.[6]

2          New hardware joins multiple super chips together to form circuits of massive neural networks comprised of tens of thousands of trillions of AI neurons (compared to 100 billion in a human brain). This enables AI to think in a way modeled after the human brain with its interconnected neurons, but so much faster and with access to so much more information.

3          The third addition to the AI mix is transformers (the T in Chat GPT) a mechanism that enables neural networks to mimic how humans learn and think by assigning “weight” or importance to each bit of new information it receives. This weighting of information mimics the human learning process. Transformers match the new information the AI receives to other similar bits of information to discern relationships and come to logical conclusions.

4          The fourth piece of the puzzle is a new ability in the processing of graphics. The same chips and processors that let video games put out such amazing graphics can be reprogrammed to take in visual information, from photos, videos, or live feeds from cameras, and learn from what they see so that AI guided humanoid general-purpose robots can learn how to perform complex tasks like cooking and serving dinner by observing you or a video. 

 Elon Musk is putting the full weight of Tesla’s engineering into the race to manufacture the first true general-purpose robot which they hope will begin mass production by the end of 2025 and will cost about $20,000 per unit. Musk seems to have two very different markets in mind for his robots based on these images released by Tesla. These two photos illustrate the double-edged nature of the promise and the peril of AI robotics.

Areas of Great Concern

The unfortunate reality is: many top experts are terrified of what unrestricted and unleashed AI will bring! It seems shocking that in spite of Musk admitting that in his opinion there is a 10-20% “chance that it (AI) will end humanity” he feels the potential benefits are worth the risk and he is guiding Tesla to concentrate it’s considerable research and development departments to AI development.[7]

AI safety researcher and director of the Cyber Security Lab at the University of Louisville, Roman Yampolskiy, puts the risk much higher at 99.99 percent![8] Most AI researchers and executives put the risk somewhere between 5 and 50 percent![9]

Microsoft founder Bill Gates had said that A.I. is potentially more dangerous than a nuclear catastrophe.

Steve Wozniak, co-founder of Apple computers, believes humans will either become pets or be wiped out altogether.

Dr. Geoffrey Hinton, the scientist who developed ChatGPT for Google, resigned in May 2023 so that he could freely speak about the dangers of AI.[10]

In March 2023 more than 1000 tech leaders called for a pause on AI, citing “profound risks to society”, including “a perfect storm of corporate irresponsibility, widespread adoption, lack of regulation and a huge number of unknowns.”[11]  The corporate irresponsibility the tech experts see is that the top AI developers – Google, Nvidia, Meta, Tesla, and Amazon  – are locked in a sort of AI arms race to develop relatively low cost super intelligent general purpose humanoid robots empowered by AI.

Despite the known risks, and in the absence of any guidance from the government, the tech giants are charging ahead, all trying to be the first with the most. Experts feel within the next six months to two years AI will be able to do everything that humans can do both mentally and physically and do it better, faster, and in the workplace – cheaper.

There are three areas of grave concern to the experts: AI driven unemployment, AI domination, and misuse of AI weaponry by governments and bad actors.

Unemployment Threat from AI: Experts agree that significant job losses will begin impacting America within five years, and will be extensive, 30% or more within ten years. The social impact caused by the elimination of 30% of jobs within 10 years will cut across multiple occupations. AI nurses, and surgeons are already being deployed, how much easier it will be to replace secretaries, factory workers, and workers in the food industry! [12] To put 30% unemployment in perspective, the unemployment rate during the height of the Great Depression in the USA was just under 25%.

AI domination: Many of us have seen movies like Terminator movies and Robocop and the Matrix movies. The reality is that negative visions of people being subjugated by machines were based on a lot of scientific truth. The ultimate fear of AI is that AI robots will analyze humans and decide we are a threat to their existence and/or the existence of the planet, and then take steps to control our population or to eliminate us all together.

I had a chat with Google’s Bard (an AI chatbot) and asked what it saw as possible threats to humanity from AI; its reply was chilling: “AI systems might be optimized for goals that conflict with human well-being. For example, an AI designed to maximize efficiency might prioritize resource allocation in ways that harm humans.”[13] In other words AI might decide that humans are a threat because we are competing with AI for the same resources!

Misuse Of AI Weaponry: You may have seen one of the drone light shows that are beginning to replace fireworks shows. Thousands of synchronized drones form beautiful images in the sky. In an article in March the Wall Street Journal envisioned the same type of coordinated drones armed with explosive charges attacking an aircraft carrier. They said AI empowered drone swarms will, “change the balance of military power.”[14] Autonomous AI drones carrying shape charges, or grenades or other weapons can be mass produced cheaper than an M-16 rifle.

Imagine a cloud of several thousand weaponized AI drones descending on a crowded stadium or a battlefield. Hollywood has already portrayed a cloud of drones programmed to kill swooping in to assassinate the president of the USA in a movie entitled Angel Has Fallen, staring Gerard Butler and Morgan Freeman.

Six Suggestions To Help Pastors and Church Leaders Prepare

Without any doubt in the short term – the next 5-7 years, AI holds many potential benefits for pastors and congregations, as the September issue of Ministry Magazine featured. AI will also provide excellent medical breakthroughs: discovering new medicines and treatments, saving lives, and lowering the cost of medical treatment and medicine and making our lives a little easier. For example, in December 2023 AI was used to compare 100,000 test results and discovered the first new antibiotic in 60 years, so important because many bacteria have become drug resistant.

However, to begin the golden age of AI for humanity that Musk and Bezos envision, our national government will need to provide extensive competent and compassionate intervention; but the likelihood of that seems very small in light of the current state of gridlock and infighting in the US Congress.

In the likely event that AI continues its current trajectory and that there is insufficient government intervention, within five years Christian churches will begin facing the twin problems of economic survival, and the need to provide pastoral care and material aid to its unemployed and unemployable members. Obviously if 30% of congregation members are unemployed the finances of the church will also most certainly suffer.

This is a situation reminiscent of Pharaoh’s dream of the seven fat and skinny cows, in that churches have five to seven “good” years ahead to prepare as best we can for the inevitable very lean years in our future. Here are six suggestions for pastors and church leaders that will help their churches to prepare.

• Find tech savvy members of your congregation who can help you keep key leaders of your church and the congregation informed about how AI is affecting your community.
• Prepare yourself and your congregation(s) spiritually for what is very likely going to be a severe time of testing.
• Form alliances with other pastors and churches in the community including other denominations, as we are reminded in scripture, “A cord of three strands is not quickly broken.”[15] Pastors and churches should be able to lean on each other for support and to form bonds that will enable them to work together.
• Set an example by putting the church in your will and personally ask every retired member to do so also. Pastors don’t like to ask for money, but this truly is a crisis we need to prepare for, so don’t be shy about strongly asking members to remember the church in their wills.
• While churches historically have distanced themselves from politics the challenges posed by AI calls for an exception; churches should exert their moral authority to petition the government for AI oversight, regulation and planning to ensure help for displaced workers, and to add national laws prohibiting the use of weaponized AI robots against civilians.
• If churches do have money in savings, carefully investing in AI stocks is a useful way to capitalize on the good years as the AI market is anticipated to grow 820% in the next five years.[16]

It is possible that the worst fears of experts may not come true. AI may be controllable and turn out to be a just blessing; but we will do well to remember that no experts see less than a 5% chance of impending disaster and many are much more pessimistic. However, church leaders should do their best to prepare themselves and the church for what is to come and trust in God for the rest.

---

[1] Gordon, Cindy. Humanoid AI Smart Robots Driving New Job Realities.forbes.com. March 18, 2024.

[2] Dumas, Breck  Nvidia announces AI-powered health care ‘agents’ that outperform nurses — and cost $9 an hour. foxbusiness.com. March 21, 2024.

[3] Gohd,Chelsea Amazon’s CEO Says We’re Living in the Golden Age of AI, Futurism.com, May 9, 2017.

[4] Mancini, Jeannine. Elon Musk Predicts A ‘Universal High Income’ As Jobs Are Phased Out And Employment Becomes Obsolete — It’ll Be ‘Somewhat Of An Equalizer. finance.yahoo.com. Mar 18, 2024.

[5] Gohd,Chelsea Amazon’s CEO Says We’re Living in the Golden Age of AI, Futurism.com, May 9, 2017.

 [6] Zewe, Adam. Massachusetts Institute of Technology.  Large language models use a surprisingly simple mechanism to retrieve some stored knowledge. techxplore.com. March 25, 2024.

[7] Okemaw, Kevin. AI safety researcher warns there’s a 99.999999% probability AI will end humanity, but Elon Musk   “conservatively” dwindles it down to 20% and says it should be explored more despite inevitable doom. Smartnews.com. April 2, 2024.

[8] Ibid.

[9] Metz, Cade and Gregory Schmidt.  Elon Musk and others call for a pause on AI citing ‘profound risks to society.’  nytimes.com. March 29. 2023.

[10] Metz, Cade. The Godfather A.I. Leaves Google and Warns of Danger Ahead. nytimes.com Updated May 4. 2023.

[11] Metz, Cade and Gregory Schmidt.  Elon Musk and others call for a pause on AI citing ‘profound risks to society.’  nytimes.com. March 29. 2023.

[12] The Future of Jobs Employment, Skills and Workforce Strategy for the Fourth Industrial Revolution. Global Challenge   Insight Report, www3.weforum.org. January 2016.

[13] Bard, Google Large language model. Conversation about benefits and risks of artificial intelligence. March 15 2024.

[14] Ackerman, Elliot  and Stavridis, James. Drone Swarms Are About To Change The Balance Of Military Power. Wsj.com. March 14, 2024.

[15] Ecclesiastes 4:12 NIV

[16] Jennewine, Trevor. The Artificial Intelligence (AI) Market Could Soar 820% by 2030: 2 AI Growth Stocks to Buy Now and Hold Long-Term. finance.yahoo.com. March 9, 2024.