Originally, the company was founded as a non-profit company with safety enshrined as its core mission to act as a check to the purely market-oriented companies seeking to develop AI. Altruism went out the window when reality hit home. The reality of how much investment is needed to hire staff, build the computing power and the data centres needed to win the race against Microsoft, Google and other companies to be the first to build AI. (It is estimated that AI-related investments in the US this year will approach $US 400 billion and that investment required by the end of the decade will be anywhere between $US 4 trillion and $US 7 trillion!)

 

The AI race is being run with no thought to the welfare of the people involved in developing AI, the writers and artists whose work is being stolen without payment to train AI, the people harmed by the misuse of AI or the people losing their jobs because of AI. There is no consideration being given to the damage done to the environment by the enormous amounts of energy needed to run AI data centres or the vast amount of pure water and land needed for them. New applications are being launched without consideration to any safety standards or protocols.

Hao writes, “Artificial Intelligence is a technology that takes many forms. It is in fact a multitude of technologies that shape-shift and evolve, not merely based on technical merit but with the ideological drives of the people who create them and the winds of hype and commercialisation.”

 

This is the heart of the matter. AI is not inherently evil. In the right hands AI can help in many beneficial ways in areas such as medicine and environmental science. Hao gives the example of how AI is being used by Maori people in New Zealand to help them preserve and recreate their language which was in danger of becoming extinct.

 

It is the misuse of AI by governments and companies that is the problem. Pornographers and scammers are using deep fakes to make millions of dollars and damage peoples’ lives. Facial-recognition technology used by police discriminates against coloured people because of the biases built in to AI in the material used to train AI systems.

 

Workers in many industries are losing their jobs because of the introduction of AI. White collar jobs in particular, are under attack in banking, insurance and government departments. Call centre workers are being sacked and replaced by chatbots. In July software company Atlassian announced that it was sacking 150 workers in customer service and support roles, with much of their work being taken over by AI. Around the same time the Commonwealth Bank announced that it was making 45 customer service roles redundant with the introduction of an AI chatbot. The bank was forced to back-track on the sackings to due to the backlash from the unions and an upsurge in calls.

 

In a speech in September, the CEO of the National Australia Bank warned that white-collar workers who don’t learn to use AI will be replaced by workers who do learn to use AI, and that there will be little or no white-collar work for those who don’t like using AI. Workers will have to embrace AI or find jobs in areas like construction and care which could not be replaced by AI. The jobs of lower-level consultants in the major consulting firms such as Deloitte are also being cut out. 

 

White-collar jobs are not the only ones under attack. The work of writers, artists and other creative people is also under threat.

 

• The jobs of musicians and others in the recording industry are under threat as companies like Spotify make increasing use of AI-generated music.

• Fashion models and photographers will lose work as fashion magazines like Vogue make more use of AI-created models in their pages.

• A Sydney radio station used an AI generated radio host for 6 months. AI-generated voices can be used to do advertisements and voice-overs, affecting many jobs in radio.

• An AI-generated actress, Tilly Norwood, is attracting a lot of attention in Hollywood. Already, screen writers and actors have staged strikes as part of a campaign against the use of AI in movie making. The jobs of the concept artists who create the characters, props and other visual elements in movies are also threatened by AI.

As the use of AI increases in film making, producers, composers, lighting people, agents and many other jobs will disappear. In October the Omni 1.0 film festival for films made using AI was held in Sydney. It attracted almost 1000 entries!

• Authors, artists, photographers, actors, musicians-all of these creative people who are so vital to our culture could vanish, replaced by AI creations. They are angry that their work is being “scraped” by the tech companies to train AI, and then appearing in AI-generated content without any payment of copyright by the tech companies.

 

Workers and all other people adversely affected by AI must unite to defeat the misuse of AI by companies and governments. AI can benefit society, but it must not be a way to make billionaires even richer!

                                                    **************************

By Alex M

 

Further to Duncan’s review of Karen Hao’s insightful and eye-opening investigation of Sam Altman’s Open AI project here are a few other things to consider about AI and the tech billionaires hell bent on driving it down our throats. 

 

First; some of the key players in developing artificial intelligence are already making fortunes out of it and potentially stand to make even greater fortunes as it develops and therefore do not have an objective view about the matter. Such a banal insight needs to be stressed because the prospect of trillions of dollars in profits is enough to open the PR floodgates in order to convince the public that AI will bring major benefits to humanity. 

 

One of the big claims made for AI is that it could eventually be employed to alleviate, if not solve the complex problem of global climate change. As Hao points out in her book, two of the leading lights of Open AI, Greg Brockman and Ilya Sutskever whom she interviewed in August 2019, were keen to impress upon her the wondrous possibilities of Artificial General Intelligence (AGI). Put simply, AGI is next level AI, which aims to match or surpass human intelligence (there are currently a number of tech companies racing to bring AGI into existence, so to speak, and these include Open AI, Google, Elon Musk’s xAI and Zuckerberg’s Meta).

 

According to Brockman and Sutskever, AGI when it is up and running will be able to hurdle over issues that stymie attempts to resolve global climate change. As identified by Sutskever, one of the foremost issues hampering resolution of complex global problems is: ‘that you have a large number of humans and they don’t communicate as fast, they don’t work as fast, they have a lot of incentive problems.’ AGI would circumvent bottlenecks such as this by scale (large numbers of computer networks) and speed of communication. (Hao, p.78)

 

When pressed by Hao that this sounded like the goal of AGI as put forward by these representatives of Open AI was to replace humans, Brockman denied it initially, before admitting that technology had always destroyed some jobs, while creating others. Open AI in its pursuit of AGI was altruistic though because it’s aim was to create AGI ‘that gave everyone “economic freedom” while allowing them to continue to “live meaningful lives” in that new reality’ . Open AI’s mission ‘is to ensure that AGI benefits all of humanity. And the way we want to do that is: build AGI and distribute its economic benefits.’ (Hao, p.79)

 

These statements give you an indication of the complete lack of insight these geniuses have when it comes to how capitalism actually works. Either that, or they are just trying to sell the product. 

 

Undercutting the big assertion that AGI would be pivotal in helping to solve the great challenge of global climate change is the fact that the data centres being constructed to house the arrays of servers needed to power AI (and the race to create AGI) are massive. So big are the data centres becoming and so resource guzzling are they that the data centres are exacerbating global warming. Power consumption is through the roof as is potable water consumption; to cool the servers requires exorbitant amounts of drinking quality water. The energy demands of the data centres are such that many tech companies have located data centres in countries in the Global South such as Chile and Uruguay where energy and water costs are relatively cheaper. 

When challenged by Hao about the problem of data centres contributing greatly to global warming via massive carbon emissions, the tech bros had no real answer. (Hao, p.80)

 

The second thing to consider is that the financing of the headlong race to develop AI and AGI has absorbed billions if not trillions of US dollars and it has generated a stock market bubble too. The major US tech corporations such as Microsoft, Google, Meta, Open AI and the computer chip manufacturer Nvidia, among others, have invested heavily in each other’s companies, creating a financial house of cards. Financial pundits have been warning of an impending bust for weeks if not months now. It may well be that if the bubble does burst, big tech corporations in the US may be deemed to be too big to fail and US pre-eminence in AI and AGI must be propped up with a bailout courtesy of the US taxpayer. Time will tell.

 

A third consideration is the geopolitical rivalry between the US and China. These two countries are in the forefront of AI and AGI development, with China making inroads into US big tech’s AI dominance in the last couple of years. The challenge from China in the field of AI has alarmed both tech billionaires and US government officials adding to anti-China frenzy and the drive to war being promoted in the US and its vassal states such as Australia.

 

A final consideration for the moment is that for all the benefits that are meant to come from AI (and possibly AGI) and there are some indeed (see Duncan’s review above) there are some decided downsides too. A prime example is the use of AI by the IDF to identify and compile targets of Palestinians in Gaza. As Duncan concluded in his review, AI can be beneficial but we must fight to control its use. It must be used in a planned and controlled way to benefit humanity as a whole, not to enrich billionaires and massive corporations.