This COVID-19 newsletter was written by a human

Summer ends in a week. I just Googled it to be sure and it still doesn’t make sense. None of 2020 makes sense. The US elections are six weeks away. It’s already pumpkin spice season. I am not ready for any of this.

I think everyone’s feeling the pain that comes from seeing life pass us by as 2020 drags on with crappy bandwidth. 

The good news should be that vaccines are imminent. We should begin to see several viable vaccine candidates emerge from phase three trials for final approval between now and early 2021. And that’s good, but it has to be taken with a bitter grain of salt. 

Vaccines aren’t off switches for diseases. We’ll actually need several vaccines to fight the coronavirus and even then, we’ll need a way to distribute them to everybody. 

The vaccine is just a small part of the solution, which is something people don’t want to hear. We’ve still got to do all the other stuff too. 

One of the biggest contributing factors to the propagation of COVID-19 has been misinformed people’s willingness to participate in the spread of partisan conspiracy theories and opine on matters they have absolutely no expertise in. 

It’s almost like millions of people are actually AIs constrained to the rules of their programmers. They have no choice, when given a specific prompt, but to generate a response based on how they’ve been trained. 

You have to admit, their behavior is more robot-like than human.

Take GPT-3 for example. It’s an AI system that was basically fed the internet as a database for generating novel text. That means it can speak on any subject… but never has a clue what it’s talking about.

Okay, now that I’m reading that paragraph again.. maybe GPT-3 is more human-like than I’ve been giving it credit for. 

This week, let’s break down the vaccine outlook. (New York Times)

Freshworks for Startups is organizing a highly curated virtual startup summit called Against All Odds for startup founders, VCs, and senior functional leaders – on September 17th.

Join 800+ founders & 20+ amazing speakers including Philippe Botteri of Accel, Tao Tao of GetYourGuide, Johnny Boufarhat of Hopin, Andrey Khusid of Miro, Anna of Zizoo, and several other leaders for inspiring hours of keynotes, on 1:1 networking, roundtable discussions, and VC connections.

Register here for the summit

No, the man who cannot be touched isn’t getting us through the pandemic (though If anyone could…). But he is one of the few people on the planet with academic access to GPT-3.

In case you’ve missed it GPT-3 is an AI system developed by Open AI, the startup Elon Musk-founded to figure out general artificial intelligence.

We’re not all that impressed with the hype surrounding GPT-3 here at TNW (I edit Neural too, where we talk about AI all day every day) but the system itself is pretty interesting.

And for me, just knowing that MC Hammer is out there testing the system and putting it through its paces brings me deep comfort.

GPT-3 isn’t “smart” at all. It’s a dumb bunch of ones and zeros that spits out the text equivalent of a hazy silhouette. For this reason, its capabilities can be instantly dismissed by any science-minded individual on the hunt for signs of sentience.

Let me be blunt: GPT-3 is no closer to human-level intelligence than a dictionary that someone’s placed a USB cord inside as a bookmark. That’s not to say it isn’t an important step on the way there. It probably is. It’s just not the “intelligent” machine the media’s hyped it to be.

But it’s incredibly interesting. We’ve never seen that much raw power put behind a text generator. The permutations that can be effectively extrapolated with billions of parameters in play result in the machine spitting out some fascinating linguistic displays.

And Hammer knows what a “fascinating linguistic display” looks like when he sees one. Who better than a lifelong wordsmith with a masterful grasp on the uniquely human penchant for turning language into art to judge the capabilities of such a machine?

This is the crossover we need.