GDP was probably a reasonable first-order approximation back in 1944, but ceased to be a good metric as soon as it became the target. The longer GDP remains our North Star, the more it becomes an increasingly hazy representation of reality. Classic Goodhart's law. I'd suggest developing a new metric that more precisely measures the things we want to optimize for. Call it something like "Real Value Product" (RVP). RVP would calculate the aggregate value of US economic output differently from GDP in a few key ways: * Filter out economic "middleware". Domestic B2B transactions wouldn’t count toward RVP, since real value only emerges when goods and services reach consumers, government, or export markets. If a B2B purchase ultimately leads to more such sales, that increase will already be captured elsewhere in RVP. If it doesn’t, it shouldn’t be counted. (Employee benefits like cars, laptops, food, and gym memberships would be exceptions, as they provide direct value to humans.) * Weight economic activity more heavily the less wealthy an entity is. This serves two functions: (1) It accounts for inequality — an economy driven by concentrated high-end consumption is weaker than one with broad-based demand. (2) It reflects spending efficiency — smaller economic units (average households, small towns) must be more selective, whereas high-net-worth individuals and the federal government have far more room in their budgets for waste. * A corollary to the above point is that building public infrastructure would count proportionally less toward RVP than actual use of public infrastructure, discouraging wasteful spending on projects that don’t deliver tangible benefits. * Correct for overpricing distortions. Instead of taking transaction amounts at face value, RVP would benchmark against global market rates where possible. For example, if a medical procedure costs $50k in the US but can be performed for $5k elsewhere under equivalent conditions, RVP reflects that only $5k of real value has been produced. (Furthermore, that $45k flowing from a middle-class individual to wealthier entities would likely depress RVP due to the inverse weighting rule.) I'd be tempted to factor in negative externalities as well, but I think that's better left to a derivative metric (something like "RVP Minus Externalities"), since our understanding of what qualifies as an externality and how to quantify its costs is constantly evolving. Implementation would likely require an AI-driven system to estimate RVP using vast amounts of public economic and financial data (potentially supplemented with anonymized IRS data if needed). @DavidSacks @elonmusk

Of course, the only solution is to defund the cartels by legalizing drugs and have the FDA regulate purity/chemistry standards. Model narcotic use reduction efforts on our highly successful tobacco policies instead of funneling money into prolonging the issue.

My solution would be: 1) gradual phaseout of politics-based weighting from the feed, 2) internal audit of moderation process for political neutrality, 3) elimination of permanent bans except where required by law, and 4) constructive feedback alongside each ban without exception.