Trusted Oracles and Bulletproof Reputations

tl;dr: Looking forward to a world where talk cannot afford to be cheap, which will make us all safer (hopefully).

Over the last few weeks, we have seen the convergence of a global pandemic, an economic meltdown, and an explosion of information.

The challenge, for each of us, is deciding what to do in reaction to these whirlwind of events.

We’ve seen rumors of shutdowns and lockdowns. We’ve heard stories of death rates and confirmed cases. We’ve seen maps from Johns Hopkins., heard stories about senators selling stock after a private briefing, and learned that police forces won’t be arresting people for many crimes.

Many of us have gotten texts like this:

We’ve heard that ibuprofen won’t work. Then, heard that “maybe it doesn’t really matter.”

We’ve heard that coronavirus only affects the old, then heard that, well, actually it does impact young people.

But very few of us actually KNOW what is true and what isn’t.

Sure, we use heuristics to assume that “well, it’s the NY Times so it must be true.”

But others say, “well, it’s the NY Times, so it must be false.”

We may like the Johns Hopkins map and trust it because of its strong brand name and reputation (and it’s my alma mater), but how do we know that the information it presents is actually true?

We can’t personally verify all 244,523 cases, but then again, neither can JHU.

as of March 20th, 4:38am EST https://coronavirus.jhu.edu/map.html

Even an institution as esteemed as Johns Hopkins has to trust people along the way in a chain of trust to provide this data.

Those people and computer systems are called “oracles” in the crypto world, but it comes down to this….how do you know that the very source of your data is 100% reliable?

Trusted Oracles

In the strange, brave, new post-corona world in which we will (hopefully) all live, this challenge is only going to get bigger and bigger.

Anyone who is on Twitter or Facebook for more than 5 seconds encounters this issue.

It’s one thing to have a verified account, the elusive “blue checkmark,” but it’s another thing to have a verified checkmark of accurate information.

Almost no one has that, but every one of us needs that.

Honestly, I don’t know how you solve the problem entirely and we certainly can’t solve it today, but I think the solution will come down to the same thing that Johns Hopkins has going for it….reputation.

Reputation as the Currency of the Future

I was introduced to the idea of reputation as currency by Matan Field, founder of DAOstack (discl: advisor).

Even though I had blogged about the topic of online reputation as far back as 2009, it was Matan who helped me elevate it from a “brand management” issue to one of “this is how you know whom to trust” issue.

It’s a big shift.

Today, or at least pre-corona (P.C.?), reputation was associated with trust primarily on the basis of reach.

CBS, CNN, Fox, NYTimes, etc….they are trusted because they won the war of awareness and distribution in their markets and maintained that position for a long time.

I’m certainly not saying that they never told the truth, but I think we all know that, at this stage of the game, sometimes the truth takes second place to ratings and sales.

And we’re the victims.

Matan saw this problem and said, “hold on, this isn’t the way to make decisions.”

It’s not about “let’s trust the people with the most money or the most power.”

Instead, it’s about “let’s trust the people who have proven themselves to be reliable and trustworthy over time.”

The partnership with Gnosis is an early example of this effort, but this is just the earliest of prototypes in what could be a seismic shift in how we allocate our trust.

Here’s a perfect example.

Talk is Cheap, Right?

I don’t know Lane Rettig. I’m sure he’s a nice guy and I have little doubt that he is trying to help.

But here’s the problem.

He’s got 11,500 followers on Twitter. Good for him. In his bio, he claims that “talk is cheap.”

Great, so far, so good.

But then comes a tweet like this.

Now, in the grand scheme of things, who really cares if Lane is right or wrong on this particular topic?

Probably no one, but….wouldn’t it be nice to know in a year or two if the pronouncements that Lane makes are accurate or way off?

The way this tweet is worded provides us with zero accountability. Just how many months and years is he talking about? What does “peak city mean?”

Would that help you in determining if you are going to listen to him the next time he makes a pronouncement?

I’m not here to pick on Lane, per se.

It was just the most recent example of this phenomenon (and I’m guilty of this as well), but if the book Superforecasters taught me anything, it’s that

“without a concrete measure and a concrete date for assessing accuracy, a prediction is totally worthless…and it’s not a prediction.”

paraphrasing idea from Superforecasters

Making Talk Less Cheap

I’ve been thinking about this a lot recently since I spent some time digging around ErasureBay from Numerai. I wrote about them before (here), but the product recently launched.

Essentially, in the world of ErasureBay, talk can’t be cheap because if you are making a claim that something is true, you actually have to stake money against it.

Yep, you may say “oh, there are 100,000,000 cases of coronavirus in China” on Twitter for free.

However, if someone on ErasureBay asks, “how many cases of coronavirus are there in China” and you say “100,000,000,” you need to back that up with real money.

Then, others can challenge your claim and, if enough people (“wisdom of the crowds” I suppose) challenge it, you lose your money.

I’m sure this method can be gamed and cheated as well in some way, but conceptually, it’s raises the stakes (literally) for lying and spreading misinformation.

Civil is trying to do this for news as well and Aragon already has a template for a reputation-based organization.

The point is this: we have to accept that we’re living in a post-truth world. It’s sad (and scary) to acknowledge that few of us know whom we can really trust on for reliable information, but the sooner we recognize that, the sooner we can start adjusting our perceptions of what is accurate and what is not.

Reputation-based systems, backed by crypto economic systems and actually staking, raising the cost of lying, offer a glimmer of hope.