In speaking to even the most critical of artificial intelligence researchers, one thing has been made very clear to me: the majority are incredibly hopeful that humanity can reach a future built around a positive technological revolution. The work they are doing, in alignment, in ethics, in sustainability, in responsibility, in transparency, is work that they undertake in an effort to achieve that future, not diminish it.
In a new post titled "The Techno-Optimist Manifesto," Marc Andreesen has branded such researchers, ethicists and scientists "the enemy."
Related: The ethics of artificial intelligence: A path toward responsible AI
The 15-section, five-thousand-word essay, published Oct. 16, makes a number of very black-and-white claims conspicuously lacking in the evidence, data and nuance that the field of AI deserves.
We are told, he says, that technology represents a threat to jobs, health, equality, the environment and society on the whole. These statements on technological risk and threat, equate, in Andreesen's mind, to pessimism about a technological future.
"We are told to denounce our birthright – our intelligence, our control over nature, our ability to build a better world," the manifesto reads.
Wow. That's a lot of enemies.
— MMitchell (@mmitchell_ai) October 16, 2023
Also sort of odd to equate people who literally bring morality to AI discourse with "demoralization". https://t.co/IVmmUFq0Fe
The threats he mentions are very real. In these early days, AI has already displaced workers and impacted fields. AI will likely broaden the gap between skilled and unskilled workers and countries, widening inequality along an exponential curve that might become unfixable, Dr. Srinivas Mukkamala told TheStreet recently. Untethered algorithms — Britain's Horizon Post Office scandal is a prime example of this — have already resulted in wrongful imprisonments and could easily result in wrongful deaths.
A chatbot in December encouraged a 21-year-old British man to kill the Queen. He is now facing nine years in prison. A Belgian man in March committed suicide after encouragement from a different chatbot. Facebook whistleblower Frances Haugen revealed in 2021 the danger of the social media giant's AI-powered algorithms, which, in an effort to keep people on the platform, push hate speech and misinformation. When she testified before Congress, she explained that "these problems are solvable. Facebook can change, but is clearly not going to do so on its own."
The risks of AI are not hypothetical. We are already living them.
And while it appears easily accessible on your phone or laptop, AI is powered by stacks of computer chips that require a tremendous amount of energy to run, thus emitting carbon dioxide and sucking up water resources at a level that is causing harm to the environment. And just like social media has increased loneliness rather than solving it, so too might evolutions of similar technology further separate people, rather than bring them together.
The mention of these negative impacts comes from a place of optimism for the future. In discussing these threats, they can be addressed, in addressing them, they can then be diminished: companies have found ways to run data centers that don't emit; others are finding ways to skill workers up; responsible AI auditing is expanding to help prevent tragic errors made by an algorithm without guardrails.
Skepticism is not the same as dismissal.
And blind faith is not the same as optimism.
Related: Artificial Intelligence is a sustainability nightmare - but it doesn't have to be
Experts take issue with Andreesen's claims
"Technology has a lot of power for both good and bad," Mukkamala told TheStreet in response to the manifesto. Focusing only on the good, he said, rather than considering the potential harms, could increase inequity, rather than solving it.
"While we should aim to be techno-optimists, we can't neglect being techno-realists and make sure that the proper guardrails are placed around the technology we create so that it truly benefits all," he said.
The broad-strokes future Andreesen, a venture capitalist worth $1.8 billion, paints — "We believe in making everyone rich, everything cheap and everything abundant" — might sound good on paper. But, according to Gary Marcus, a leading AI researcher, such a future is achievable by working to recognize and address its pitfalls, risks and threats, hypothetical or painfully real.
I can empathize with Andreessen's creed to a degree — let's not unnecessarily halt the advance of technology. But as a father of 5 digital natives, 4 of which are now adult age, unfettered technology is a recipe for regret, remorse, and ruin.
— Cameron Moll (@cameronmoll) October 17, 2023
Reaching those positive outcomes, Marcus wrote, means "working for a positive future, not just declaring that one is inevitable and hoping for the best."
The enemy of tech
Andreesen goes on to claim that such movements as "sustainability," "social responsibility," "trust and safety," "tech ethics" and "risk management" are all part of a "mass demoralization campaign" that has been ongoing for 60 years. The people who belong to such categories of thought, he claims, are "our enemies."
Without such groups, though, any benefits that can be derived from said technology will benefit only the slimmest group of people. And they will disenfranchise the rest.
"Technology can liberate on one hand, and shackle on another," Nell Watson, a leading AI researcher and tech ethicist told TheStreet.
"We must be ever mindful of unanticipated social effects of technology, as it takes society in strange new directions. All that glitters is not gold."
AI, properly handled and leveraged, can enhance the fields of science and medicine, allowing us to quickly unlock solutions that were previously out of reach. But as Suresh Venkatasubramanian, an AI researcher and professor who in 2021 served as a White House tech advisor, told TheStreet in September, the way forward is not through halting innovation, it is through powering responsible innovation.
I couldn’t decide whether I should laugh or cry…
— Dorothea Baur (Dr.) (also on bsky) (@DorotheaBaur) October 16, 2023
Laws, he said, that enforce ethical and trustworthy AI systems, will create a market demand for responsible AI innovations — which are already happening — that will ensure models are not only safer but more usable. This growing demand for ensuring the responsible use of such technologies will shift the educational process, a pathway that is in no way preventative of a technologically-minded future.
It is just more cautious. More careful. More inclusive.
"A measure of optimism is essential for innovation. We have to believe that it’s possible to change the world for the better, in order to dare to try, fail, and try again," Watson said.
"However, optimism needs to be balanced with rational caution. Pathological optimism in the face of genuine peril can be toxic, even catastrophic."
Related: Huge new ChatGPT update highlights the dangers of AI hype
Get investment guidance from trusted portfolio managers without the management fees. Sign up for Action Alerts PLUS now.
