Game of Tweets | NFT Bubble | The AI Mind
Crafted while realizing that no one knows what the future holds
This gentle orchid is a tough customer (Credit: Lance Ulanoff)
Game of Tweets
I’m exhausted by this whole Elon Musk Twitter mess. It’s gone well past being intense, worrisome, important, and is now bordering on ridiculous. Billionaire Elon Musk spends most of his time tweeting and undermining whatever cachet he’ll have if and when he does own the company.
Questioning the company’s metrics and bot numbers is fine. Doing it publicly is counterproductive. I also wonder if current CEO Parag Agrawal now sees an opening, a way to wedge Musk off his Twitter acquisition hopes. Initially, I thought Agrawal was on board, but now it’s clear that he sees Musk as a hostile entity and would the Tesla CEO doesn’t buy the company. Agrawal’s clear, tweeted explanations - while keeping some of the company’s proprietary information secret - are designed, I think, to inform, but also paint him as a rational leader while drawing Musk in to tweet semi-irrational responses.
It all just seems to weaken Musk’s position day by day.
The problem is that Twitter may not be the same company when all is said and done. If Musk doesn’t succeed, it is not necessarily good news for Twitter. It’s been deeply harmed, weakened, and will take some time to right and re-energize itself.
For all Musk’s bluster, he does have some interesting ideas. I wish he had worked to become an active shareholder and not an owner so he and Agrawal could actually work together to test some of these concepts. If or when Musk gets in, I think he may just ram his own ideas through.
I do not have a good feeling about all this.
NFT Bubble
I have nothing against non-fungible tokens and the concept that this is finally a way for artists to maintain control and ownership and get paid for their work.
I call it a “concept,” because, despite NFT’s attachment to the immutable crypto ledger, nothing about them feels real, safe, or permanent.
And yet, so many industries are flooding into launching their own NFTs. Sports teams, celebrities, sneakers, tech companies, gaming…honestly, the list is endless.
In recent weeks, huge NFT marketplaces, including one of the biggest, Bored Ape, have been hacked, and others have had their collections stolen from them. Scammers recently stole $300k worth of NFTs from actor Seth Green.
Earlier this year, Open Sea reported that 80% of the freely created NFTs on its platform were frauds or spam.
Right now, the NFT market is worth almost $45B. That number alone should scare most people because it’s a ton of capital sunk into a wildly unproven investment. The rapid growth of NFTs, people’s lack of full understanding of the mechanisms behind them, and their obvious vulnerability to hackers and scammers should be giving people pause.
It’s not, obviously, and as I tweeted this week, this moment feels a lot like the years and months leading up to the dot-com bubble and burst. So much money was invested because there was no down, only unlimited opportunity.
In a way, there were right. The internet has transformed business, finance, and society, but it didn’t do so before a major correction. I think NFTs and, yes, cryptos are primed for the same sort of adjustment. In fact, it might be happening right now.
Computer Mind
Are we thisclose to a sentient computer? Is it “game over” for humans? Maybe, maybe not.
When Google’s Deep Mind research Nando d Freitas posted his “opinion” that, in regard to AI achieving Artificial General Intelligence or “the ability of an intelligent agent to understand or learn any intellectual task that a human being can” it was “Game Over” people took that to mean that researchers had achieved human-level reasoning in an AI.
A more careful reading of his Twitter thread makes it clear that while we’re closer than ever, we’re still, as he noted, missing the scale, computational efficiency, and smart enough memory. More importantly, they still must build the right models.
The message is that human-level computer intelligence is not here yet, but they’re proving it’s attainable. We will solve that scale problem, we will train up the models, and build stronger algorithms, neural networks, and machine learning.
The real takeaway here should be that we better get ready and make sure that we are not building AIs that act like us but lack rational unbiased thought. If we don’t weed out biases and program in a moral compass now, we are doomed.