News: Amazon blocks Meta’s AI agent, Ryft raises £20m, and Sprive bags $10m

Laura Watkins (0:04)

This is Fintech Insider News. This week, Amazon blocks Meta's Muse AI agent, Ryft raises £20 million series B for international expansion and AI-driven cashback to mortgage app Sprive raises $10 million.

We'll be tackling all of this and more on today's news show, so don't go anywhere.

Hello, and welcome to Episode 1102 of Fintech Insider News, brought to you by 11FS, the sixth time consultancy of the year that works with financial providers big and small to build the next generation of financial services. I'm Laura Watkins, Director of Media and Marketing here at 11FS. And to help me unpack the biggest and most interesting stories from Fintech and financial services from the past week, I'm joined by a brilliant panel of guests. First up, we have a welcome back to the podcast for Sadra Hosseini, CEO at Ryft. Welcome to the show and congratulations on your big news. We'll talk about the details very shortly, but can you introduce our listeners to you and to Ryft?

Sadra Hosseini (1:18)

Good afternoon. Thank you so much for having me. Great to meet yourself, Oban and Shareq as well, on the call.

I'm Sadra, CEO and co-founder of Ryft. Ryft is the payments infrastructure for multi-party commerce. What does that mean is marketplaces, digital platforms, franchises, anywhere where there's a need for one too many or many too many transactions in the payment flows.

Laura Watkins (1:41)

Amazing. Well, thank you so much for joining us.

We also have a welcome back to the show for Oban MacTavish, CEO and co-founder at Spade. Welcome back. Can you remind us a little bit about yourself and Spade?

Oban MacTavish (1:53)

Of course. Thanks for having me.

At Spade, we build data infrastructure for financial services and banks, and we help them create value from their large data asset, their payments data. Really what that means is we ingest low-quality payments data. We help them get better in real time and help them better understand where their customers spend their money.

Laura Watkins (2:10)

Wonderful. I look forward to getting your insights on today's stories.

Last but not least, we have Spade Shareq Husain, Chief Operating Officer at Just Move In. Welcome to the podcast. Can you please tell us a little bit more about yourself and Just Move In, please?

Shareq Husain (2:24)

Yeah. I lead product and strategy at Just Move In. Just Move In is essentially a marketplace for home services meets OpenClaw. It's an agentic platform that manages everything for your move, and then everything once you've moved in to manage your home, which is where people spend the most time and money.

Laura Watkins (2:43)

Wonderful. Well, thank you so much for joining us, everybody. Without further ado, let's jump in to our first story, which is that Amazon has blocked Meta's Muse AI agent. This story on Fenextra and many other places. Amazon has blocked Meta's new Muse AI agent from shopping on behalf of customers on its platform, saying that the agent is not authorized to access the site. Amazon says Meta did not notify it that Muse would access Amazon or agree to request to keep the agent off the platform. According to GeekWire, Muse does not identify itself while browsing and appears to capture and store customer credentials raising privacy and security concerns. Muse launched earlier this month as Meta's personal AI agent and can perform tasks on users' behalf including checkout and payments through a partnership with Stripe. Amazon has previously restricted agents from Google and OpenAI and is currently in a legal dispute with perplexity over its Comet browser.

Wow. Okay. So there's a lot of moving parts here. Omer, maybe you can start us off. Amazon's argument is essentially that if an AI agent wants to transact on its platform, Amazon wants to know about it and have a say on whether it should be there essentially.

Is that a reasonable expectation from them? What's your take on this?

Oban MacTavish (3:58)

I think it's absolutely reasonable. I mean, at the heart of this, if the agent was just browsing, I don't think Amazon would have a problem with it. The moment you're acting on behalf of a user and there's a payment involved, the merchant is taking on risk.

If you don't even know it's an agent, you don't know how to properly risk rate this transaction. If I have an Amazon account, they have a perception of what I bought, they've seen my credit card before, they understand my behaviors. I think there's a real question that says, if you're like an agent acting on your behalf, who does bear the risk? Amazon is certainly not going to be interested in having all the disputes that might happen. If my agent buys the wrong thing, I start filing a dispute. The issuer is upset, Amazon is upset, everyone is upset. And Meta is very likely not holding the bag, which I think is part of the risk with all these frameworks that exist today around agents. I think it's completely valid to say, we need to know this is an agent so we can properly assess if we want to let this transaction through because once you're moving money, I think risk becomes a very big question.