The platforms you build on just changed the rules

The companies you plug into for payments, distribution, and infrastructure moved a lot of pieces this month. A new bank inside a social app. A $53 billion payments bid. Two lawsuits about who owns your people and your ideas. A court fight over whether the government can blacklist a vendor for talking back. None of these is your story alone, but together they change how you should think about the partners you lean on.

Let me catch you up.

The everything app wants your checkout

X flipped on a full bank inside the app. X Money now offers peer-to-peer payments, a Visa debit card, direct deposit, bill pay, and up to 6% APY for Premium Plus users. Deposits sit with Cross River Bank, insured to $250,000. This is Elon Musk chasing the "everything app" he has wanted since he ran a payments company called X that merged with PayPal.

Whether X Money survives matters less than the pattern. Big platforms want to own the money layer, not just the feed. The Verge noted Senator Elizabeth Warren already flagged risks to "consumers, our national security, and the stability of the financial system." If your product sits inside someone else's app, assume they will eventually want a cut of the transaction, not just the attention.

Your payments vendor might get bought

Stripe and Advent put $60.50 a share on the table for PayPal. CEO Enrique Lores did not slam the door. He said PayPal would "carefully consider" any path that created "superior value" for shareholders, per TechCrunch. Cantor pegged fair value closer to $70. So the answer is not no, it is not at that number.

If PayPal or Venmo is in your checkout, a Stripe deal would reshape your options over the next year or two. Roadmaps freeze during M&A. Support gets thin. Pricing gets revisited. You do not need to panic, but you do need a second payment rail you could switch to without a fire drill.

Contracts stopped being enough

Two suits show how loose the rules got. Warner Bros. Discovery sued Amazon for poaching an HBO Max marketing exec 16 months before her contract ended, calling Amazon a "digital bull in a china shop." WBD wants a court order blocking Amazon from hiring any of its contracted staff. The complaint says the exec's exit told the ranks that deals can be tossed "whenever a larger paycheck appears."

The other one hits closer to product. Runlayer, an MCP gateway startup, says it walked Rippling through nearly a year of trial, sharing roadmap and source code under an NDA. When they could not agree on price, Runlayer claims a Rippling insider texted that there was "a project internally to build essentially a clone," "almost a 1 to 1 copy." Rippling denies it and says it built a superior product on its own information.

Selling to a company that can build it

Runlayer's problem is the one facing anyone selling infrastructure to tech buyers. The customer has the engineers to make the thing themselves. A deep, hands-on trial is how enterprise deals close, and it is also how you hand over your blueprint. An NDA and a no-copy clause are what you are left holding if the buyer walks and ships a lookalike.

That does not mean stop selling. It means protect the crown jewels. Stage what you expose during a trial. Keep source code out of it as long as you can. Price to close, because a stalled deal with a big engineering shop is the exact spot where they decide to build instead of buy.

When your customer can punish you for talking

The Anthropic case is the one to sit with. Anthropic told the Pentagon it did not want its AI used for mass surveillance or for lethal targeting decisions. The government tried to brand it a "supply chain risk" and ban it across federal use. One argument: Anthropic's public criticism of the DoD justified the ban. Judge Rita Lin called that logic "really troubling" and found no proof Anthropic could flip "some kind of kill switch."

A big customer tried to blacklist a vendor for saying no. The court pushed back, but the attempt is the signal. Your biggest account has more leverage than the contract shows, and they may use it when you set a boundary they do not like.

The deep cut

The common thread is concentration. The platform that distributes you also wants your payment layer. The partner you trial can clone you. The customer you depend on can try to blacklist you. Every one of these stories is a company with more power than its counterpart, testing how far that power goes.

So do the boring homework before your next review. List your top three platform dependencies: payments, distribution, key vendors. For each, write down what happens if that partner gets acquired, decides to compete, or turns on you. If the honest answer is "we're stuck," that is your Q3 work, not a someday problem. Peacock's YouTube bundle deal shows the upside of spreading your bets: it put content on Amazon, Apple, and now YouTube, and posted its first quarterly profit with 48 million subscribers. Distribution in many places beats depending on one.

Three questions for your team

• If our main payments provider got acquired tomorrow, how fast could we switch, and what breaks if we can't?
• During enterprise trials, what exactly are we exposing, and could a buyer rebuild our product from what we hand over?
• Which single partner or customer could hurt us if we told them no, and what would reduce that dependence this quarter?