E112

E112: Is Davos a grift? Plus: globalist mishaps, debt ceilings, TikTok's endgame & more

2023-01-20 spoken.md · speaker-labeled ▶ watch ← E111 all episodes E113 →

5
ideas born
8
ideas moved
15
captures · 4 voices
4
dissenting
+184.8
conviction added
-129.1
decay · 110 silent

Every number here is replayed from score_events — the same ledger the pool ranks on. Decay is what the 110 ideas nobody mentioned gave up this week; it applies only when an episode is processed.

Tier crossings

conviction thresholds crossed by this episode — 65 / 45 / 15 · ideas born here show where they landed. why 65 isn't always green

▲ +55.3 🏛️ TikTok crackdown impairs ByteDance and hands share back to US platforms born at watch 55.3
▲ +42.9 🏛️ Debt-ceiling standoff resolves — they raise it and keep spending born at ember 42.9
▲ +28.8 🛢️ Under-surveyed resource geographies reprice as supply chains route around China born at ember 28.8
▲ +26.1 ⚡ China's nuclear build gives it a cheap-electricity manufacturing edge born at ember 26.1
▲ +25.0 🤖 AI drug discovery hasn't lifted pharma hit rates — the M&A bid is falling born at ember 25.0

Kill dates that landed since E111

0 hit · 0 partial · 1 miss — windows that closed after 2023-01-13 and up to 2023-01-20, auto-scored against price data and never hand-set. verdict · R · α

ideaverdictRαclosed
📈 EBV-causes-MS proof turns Epstein-Barr into a funded therapeutic target MISS -73.9% -62.6 2023-01-15

Who moved the board

each voice's force on conviction this episodesupports and opposes, weighted exactly as the replay applied them · share = % of this episode's movement

Chamath
Chamath
5 captures · 59% of movement · 4 ideas born
+152.5 / -6.5 → net +146.0
Friedberg
Friedberg
4 captures · 23% of movement · 1 idea born
+37.9 / -23.0 → net +14.9
Jason
Jason
3 captures · 11% of movement
+15.9 / -12.6 → net +3.2
Sacks
Sacks
3 captures · 8% of movement
+20.7 → net +20.7

What got argued (8 ideas)

ordered by how hard each idea moved · quotes are verbatim from the transcript, timestamps deep-link into the episode

NEW META 🏛️ TikTok crackdown impairs ByteDance and hands share back to US platforms closed 11 CONTESTED ▲ +55.3 0.0 → 55.3

All the political anger aimed at Facebook and Google gets redirected at ByteDance because picking a fight with a Chinese company is easy and popular, and US big tech will happily point the finger over there. Chamath's read: TikTok's enterprise value is badly impaired (a $70-80B markdown, 35-40% off the $320B mark), the next shoe is advertisers pulling back under political pressure, and cap-table holders should sell into secondary. ByteDance is private, so the tradeable side is the US ad platforms that absorb the share and the ad dollars.

plays META ·primary GOOGL SNAP evals 2024-01-20
Chamath
Chamath support ×3 explicit_prediction ▶ 1:02:44
So I don't think this is going to end well for TikTok. And I think the goal, if I were any of these people on the cap table, would be to sell it in secondary to somebody else and get out. I think the next big shoe to drop is going to be advertisers who come under a lot of pressure.
Jason
Jason support ×2 sentiment ▶ 1:05:05
This is the same as the 5G issue. You cannot trust the Chinese government to not steal intellectual property or to put back doors into the software. It is common business practice there. Huawei was banned.
Sacks
Sacks support ×2 explicit_prediction ▶ 1:06:54
I agree with Chamath that the future is not bright for TikTok here because it's gotten caught up in the geopolitical rivalry between China and the United States, and that's only going to keep getting more and more intense.
Friedberg
Friedberg oppose ×2 sentiment ▶ 1:14:21
I mean, look, guys, simpler things fail. I don't know how you turn off TikTok for 100 million people that are using it for two hours a day.
NEW SPY 🏛️ Debt-ceiling standoff resolves — they raise it and keep spending closed 13 CONTESTED ▲ +42.9 0.0 → 42.9

Chamath's call on the 2023 budget showdown: debt-to-GDP has no magic failure threshold, so the fiscal hawks lose. Republicans hem and haw and then capitulate, the ceiling gets raised, and Washington keeps funding the onshoring/deglobalization build-out because that spending lands in red states that would otherwise wither. No default, no forced austerity — a non-event for risk assets.

plays SPY ·primary PAVE TLT evals 2024-01-20
Chamath
Chamath support ×3 explicit_prediction ▶ 40:37
And in that context, I think that people will him and ha, but ultimately, they'll capitulate, they will raise the debt ceiling, and they'll continue to fund this transition away from globalism.
NEW INDA 🛢️ Under-surveyed resource geographies reprice as supply chains route around China closed 11 ▲ +28.8 0.0 → 28.8

Chamath's disclosed positioning: he is putting real money into places that control natural resources nobody has properly mapped — India above all, where geological survey coverage is naive relative to what's there, plus Indonesia, Australia and parts of Africa. Governments in those places are sophisticated enough to price full employment into their cost of capital and accept a longer payback to keep processing at home rather than shipping inputs to China, so the raw-material rents move out of China's orbit even before energy costs are considered.

plays INDA ·primary EWA PICK evals 2024-01-20
Chamath
Chamath support ×3 positioning ▶ 47:33
I am investing a lot of money in those places that control the natural resources that are poorly understood, meaning, there are places like India where our geological survey capacity is relatively naive.
NEW URA ⚡ China's nuclear build gives it a cheap-electricity manufacturing edge closed 52 CONTESTED ▲ +26.1 0.0 → 26.1

Friedberg's repeated claim, framed as playing out over the next couple of years: China is building 450 nuclear power plants, driving industrial power below 4-5 cents/kWh and electrifying its factories, so the unit of production stops being human labor and becomes a unit of electricity. Whoever has the cheapest electricity has a structural manufacturing cost advantage, which undercuts the West's onshoring economics. Chamath disputes it — cheap energy doesn't help if you don't control the material inputs.

plays URA ·primary COPX FXI evals 2025-01-20
Friedberg
Friedberg support ×3 explicit_prediction 24mo horizon ▶ 44:24
My point was that in China, and I've made this point many times, but I just think it's a really important one that will play out over the next couple of years. They're building 450 nuclear power plants, they're going to get the cost of industrial power below five cents or four cents a kilowatt hour, and they're electrifying all their factories.
Chamath
Chamath oppose ×2 sentiment ▶ 46:09
Even if energy is zero in China, you have to think about inputs, meaning factories make things with inputs. And if you look, for example, in natural resources, the inputs, by and large, don't exist in China.
NEW RXRX 🤖 AI drug discovery hasn't lifted pharma hit rates — the M&A bid is falling closed 4 CONTESTED ▲ +25.0 0.0 → 25.0

Chamath's contrarian read off the BioNTech/InstaDeep deal: exit values for ML-in-pharma companies have gone DOWN even as the technology got far better (Flatiron Health sold for $1.9B in 2018 and is still the high-water mark), because none of it has been shown to improve the design-space guess that actually drives pharma's batting average. Every pharma gets the capability as an adjunct and pays less for it, so the listed AI-drug-discovery cohort is overvalued; the only model that captures real value is giving the tools away for a back-end royalty (Royalty Pharma is his template).

plays RXRX ·primary SDGR evals 2024-01-20
Friedberg
Friedberg oppose ×2 explicit_prediction ▶ 1:17:22
So if you take 99% of the junk out of the top of your pipeline and you only focus on the 1% that the software predicts will be more successful, you much more quickly get drug discovery through the pipeline and you have a much higher hit rate.
Chamath
Chamath support ×3 explicit_prediction ▶ 1:25:52
And so what's really happened is the value of acquisition in M&A has gone down, even as the technology capability has gone way, way up. And why is that? It's because this stuff has yet to be proven to actually meaningfully improve the hit rate for these drug companies.
GOOGL 🤖 Natural-language chat disrupts Google's search box closed 0 CONTESTED ▼ -12.6 64.8 → 52.2
Jason
Jason oppose ×2 sentiment ▶ 1:28:15
Now I'm in like, that phase where I'm like, yeah, this is impressive, but it didn't actually solve my problem. And it's slightly faster than doing a Google search.
AAPL 🤖 Models commoditize — proprietary data is the only AI moat closed 4 CONTESTED ▲ +10.8 73.1 → 83.8
Friedberg
Friedberg support ×3 explicit_prediction ▶ 1:17:22
When you can have large, unique data sets that you can then model using these tools and these capabilities and be predictive about what the next product iteration should be, it can really change the value and the trajectory of your business.
Sacks
Sacks support ×2 positioning ▶ 1:21:27
Where it's really powerful is over time, right? If it's got your last six sets of x-rays over a, whatever, six-year period, it can detect changes that are probably, you know, hard for a human to see.
Jason
Jason support ×2 explicit_prediction ▶ 1:28:15
So I am thinking there's going to be a really good business created in taking the open source projects and forking them and verticalizing them like, you know, Sacks's one that's doing dental work, you know, like this makes sense to me.
NVDA 🤖 Generative AI is the next Silicon Valley bubble cycle closed 24 CONTESTED ▲ +8.5 26.6 → 35.1
Sacks
Sacks support ×2 sentiment ▶ 1:23:08
AI is already the hot thing. Everyone's kind of looking at this now.

Episode digest

written during extraction and stored in data/extractions/ep112.json — the auditable source of truth, including everything market-adjacent that did not earn a capture

Friedberg guest-moderated (Jason lost his voice and sat in the guest chair) and the fiscal thread was the episode: Friedberg restated the debt burden as his number one concern on Earth, Jason publicly converted to austerity and backed the Republicans in the debt-ceiling fight, and Sacks cited 350% global debt-to-GDP — three voices reinforcing the E024 fiscal-crunch idea, while Chamath took the other side hard, calling debt-to-GDP an intellectual red herring with no magic failure number and predicting Congress capitulates, raises the ceiling and keeps funding the onshoring build (new idea). Under the Davos grift-commentary, Chamath and Friedberg both reaffirmed deglobalization as structurally higher prices (Chamath again rejecting Friedberg's lower-growth rider), Sacks pinned China's rise on MFN and a coming security competition, and Chamath disclosed he is deploying real money into under-surveyed resource geographies, India above all. TikTok drew the sharpest new call: Chamath says ByteDance equity is impaired 35-40% off its $320B mark, advertisers are the next shoe and holders should dump into secondary — Sacks and Jason agreed, Friedberg dissented that you cannot switch off an app 100 million people use two hours a day. On AI, Chamath argued ML-in-pharma exit values are falling because nothing has improved pharma hit rates (Friedberg disagreed via the in-silico funnel argument) and Jason called ChatGPT a parlor trick only slightly faster than a Google search. DIARIZATION: there is no David Friedberg label in this file — his 106 turns are labelled 'Nick Friedberg', the producer's first name welded onto Friedberg's cluster (same shape as 'Erik Friedberg' in E093/E095). Content confirms Friedberg throughout (Google 2004, Monsanto, ag/pharma ML, the Yamanaka epigenetics science-corner sign-off naming his three co-hosts, addressed as 'Friedberg' by name), so all his mentions are canonicalized to Friedberg; the other three labels content-check clean.