98% uptime is not great. Our eng department is thinking about going half half with Codex but of course there’s a switching cost.

I'm VERY curious about your case. What kind of switching costs do you guys have? I'm working at a very young startup that is still not locked into either AI provider harnesses -- what causes switching costs, just the subscription leftovers or something else?

subscription leftovers are noise. the real switching cost is the harness glue.

prompts. tool calling quirks. evals. auth. retries. all the weird failure modes your team already paid to learn.

FYI I use my Codex models with Claude code and they work pretty great. It can even pick up on existing conversations w/ Opus and then resume w/ OAI models.

I answered 8/10 correctly but mostly on instinct, for example betting that the Trump tweet is misleading. Opus 4.6 got 9/10 correct. You might need an internal time limit (don't show the user) and some strawberry questions.

In the case of Iraq, they lied on purpose to support the invasion. In the case of Iran, Trump just ignored the intelligence. I do think the intelligence community is capable. For example, they warned of the Russian invasion weeks before it happened when all other European countries said it wouldn’t.

> all other European countries said it wouldn’t

This is false, France made a misstep and even fired the head of French military intelligence because of the failure to predict the invasion and Germany was also notably skeptical. Other countries like UK, Poland, Estonia, Latvia, Lithuania were on board that the invasion was going to happen.

On the Strait of Hormuz and surrounds they were capable.

Not always the case though: https://www.everycrsreport.com/reports/98-672.html

I think also the actual Chernobyl fallout might have something to do with it. They just declared wild mushrooms safe to eat last year, 40 years later.

Of course I agree with you though. This fear based approach doesn’t matter when France, Switzerland and Belgium have tens of reactors on the border.

It’s a tale as old as time: start a war to support the military industrial complex. Imagine a $4 billion investment into public transportation or parks. Every 10 years we can invest into a new city instead of bombing some kids overseas (whose siblings, fueled by hatred, then commit terror attacks on the west).

I doubt that good looking IMAX quality astrophotography is just a prompt away.

The United States is a bully in a China shop and that China shop is the world. It constantly destabilizes the Middle East under false pretenses and lets Europe pick up the refugees. It brain drains and takes innovation to be commercialized. It makes money on the hard labor of others. So yes probably for the best for Europe to do what China is doing and decouple.

Specific examples: Francis Lowell stealing British Loom designs in 1800.

Bessemer steel process replicas. Bell vs European Telephone patents.

Radio: Marconi was initially backed in the US then Bell filed the patents.

Operation Paperclip and all that entails (thousands of German patents invalidated and filed in the US).

They are reprogramming a US built drone to the German datalink equivalent with some AI sprinkled on top. Unfortunately far away from a real industry.

That's how it was always done. Nobody invented the whole thing from scratch.

Some other country started by _stealing_ US tech and now they arguably have a higher military tech throughput. Everyone has to start somewhere..

At this point purchasing Google services/products is a real risk for business continuity.

It really is. You could not pay me to tie my business to Google at this point. I need someone I can trust won't just pull the plug in a year when they get bored, and Google isn't that company.

They sold off Google Domains to Squarespace a year or so ago which was irritating because it was super nice to grant DNS record access using employees' Workspace accounts.