I’m going to confess that I was unaware that Marc Andreesen, formerly a software developer and now a venture capitalist, had written a lengthy manifesto until Adam Lashinsky pointed it out in his Washington Post column. I don’t plan on reading Mr. Andreesen’s 5,000 word defense of “techno-optimism”. I have other demands on my time. Having read Mr. Lashinsky’s column I’m somewhat at a loss at his negative reaction. Does he have something personal against Mr. Andreesen? Or is it just that he makes money?
Here’s the kernel of the column:
Andreessen doesn’t share his views for the sheer joy of airing them out, of course. He’s clearly concerned that lawmakers will put guardrails on artificial intelligence (which they should). “We believe any deceleration of AI will cost lives,†he writes, deploying a royal “we†throughout to encompass other techno-enthusiasts who share his point of view. “Deaths that were preventable by the AI that was prevented from existing is a form of murder.†In his view, AI is no danger; it is here to save us from ourselves.
Another agenda is at play here, of course. Not everything Andreessen Horowitz touches turns to gold. The firm has been a major investor in cryptocurrencies and stands to lose millions from that field’s implosion. Its investment in Twitter at a $44 billion valuation will likely never pay off. Andreessen might just need a win. And he isn’t above putting his thumb on the scale to oppose sorely needed regulation.
I suspect my own hobbyhorse is somewhat different from either Mr. Andreesen’s of Mr. Lashinsky’s. I think that wealth, size, and power, an intimately related threesome, all need guardrails and that pertains not just to companies but to individuals, governments, and all organizations.
Focusing on corporations for a moment, my observation has been that they become wealthy, large, and powerful by responding to the environment. They make something or perform a service that is wanted. They expand and retain their wealth, size, and power by manipulating the environment, particularly the legal and political environments.
I don’t really understand Mr. Lashinsky’s view of artificial intelligence. My view is that constraining it is impossible. Once released the jinn cannot be returned to the bottle. He exhibits the viewpoint of those who believe that any ill may be eradicated by passing a law against it which is clearly untrue. That’s why, for example, despite four millennia (at least) of laws against murder, there are still murders.
