A library of snippets
xxx
While a large number of performers on sites like OnlyFans make content from home, the civil servant, who refers to herself as Katie, appeared to be making adult content alongside her purported regular day job.
In a video seen by The Sun, she can be heard giving advice to what she claims is someone calling about their pension
xxx
xxx
The word “agent” conjures images of autonomy, but agentic software is not autonomous in the strong sense. It’s delegated intelligence within a larger deterministic system.
From: What Agentic Software Really Means: Beyond LLM Wrappers and Hype.
xxx
Pix is a phenomenon.
This concentration of power in a central bank is unusual, and has led to criticism. “Now we live in a democracy, but imagine if this existed under an autocracy and all your information was available to the government,” says the head of one prominent fintech company. He thinks citizens in richer countries would balk at the government having Pix’s level of access to all financial transactions. Also, if the system is ever hacked or breaks down, the fallout would be greater than if a single bank were attacked.
From: Brazil’s government-run payments system has become dominant.
This may well be one reason why the Brazilian central bank is so keen on Central Bank Digital Currency (CBDC)
xxx
the Drex Platform is a Distributed Ledger Technology (DLT) ecosystem, in which regulated financial intermediaries will convert balances of demand deposits and electronic money in Drex, so that their clients have access to various intelligent financial services.
xxx
xxx
The system is structured in two instances: wDrex (wholesale Drex), which facilitates interbank transactions and settlements, and rDrex (retail Drex), which enables everyday payments in the form of peer-to-peer transfers. By leveraging blockchain’s efficiency within a secure and regulated framework, Drex aims to reduce transaction costs, foster innovation, and create a more inclusive financial system in Brazil.
xxx
Many CBDCs worldwide are designed primarily to function as digital cash, making payments faster and more efficient. Drex, however, is different. It is built to expand financial inclusion by introducing tokenized assets and smart contracts—two features that could democratize credit access and investment opportunities.
One of the most promising applications of Drex is tokenized lending, which could provide small businesses and individuals with new forms of credit. By allowing assets to be tokenized—whether real estate, government bonds, or other securities—Drex makes it easier for borrowers to access collateralized loans in a more efficient, transparent manner.
At the time fo writing, Drex appears to have hit something of a roadblock with respect to privacy.
A working paper from the Brazlian central bank looked into this in some detail and they found that while a bank’s customers benefit from instant payments there is a corresponding impact on the bank’s’ ability to manage the timing of their
payment flows. The reduced ability to delay and net off of payment flows means bank are more exposed to shocks and therefore they hold a bigger propotion of liquid assets. In effect, banks become “narrower”.
xxx
Turkle reminded the audience of something painfully simple, that we are vulnerable to things that seem like people.
Even if the chatbot says it isn’t real, even if we rationally know it’s not conscious, our emotional selves respond as if it were. That’s how we’re wired. We project, and we anthropomorphize to connect.
From: The Human Cost Of Talking To Machines: Can A Chatbot Really Care?.
xxx
xxx
While everyone is rightly excited about the invention of the agentic AI equivalent of the /cgi-bin–Model Context Protocol (MCP) or Agent-to-Agent (A2A) communication, we are repeating the mistakes of the dot-com era–implementing the capability before we understand how to secure it.
From: (25) Without a Local Policy Engine, Your AI Agent Is a Rogue Intern | LinkedIn.
xxx
xxx
On Monday, a developer using the popular AI-powered code editor Cursor noticed something strange: Switching between machines instantly logged them out, breaking a common workflow for programmers who use multiple devices. When the user contacted Cursor support, an agent named “Sam” told them it was expected behavior under a new policy. But no such policy existed, and Sam was a bot. The AI model made the policy up, sparking a wave of complaints and cancellation threats documented on Hacker News and Reddit.
From: Company apologizes after AI support agent invents policy that causes user uproar – Ars Technica.
xxx
xxx
EU capitals reached an agreement on the Commission’s draft law in December, paving the way for discussions with EU lawmakers to make the proposals a reality. They said the “better data sharing” rules would enable financial players to target consumers with “highly personalised financial products and services.”
But financial lobbyists heavily criticized the proposed rules. In December, six industry associations, representing banks, insurers and asset managers, argued the Commission hadn’t adequately calculated the cost of the proposed rules and that customer and market demand for such data-driven services had not been proven. The group said various key concerns “repeatedly raised” remained “largely unaddressed.”From: Brussels U-turns on plans for more consumer financial data access – POLITICO.
xxx
xxx
It’s hard to say precisely how a broad-based digital verification system would work
From: Tokenization of the market coming if we fix one problem: BlackRock CEO.
xxx
Tobi Lütke, CEO of the Canadian e-commerce player Shopify, instructed staff to prove that a job can’t be done using AI before requesting new staff or resources. Since 2022 the company has cut its workforce by a third, which makes it an intereting case study for fintechs developing their AI strategies.
Swedish fintech Klarna is also an interesting case study on the deployment of AI. They were also early advocates for the new technology and according to Sebastian Siemiatkowski, Klarna’s CEO, their sales and marketing expenses in the first haf of last year fell 16% compared with the same period in 2023, while customer service and operations expenses shrank 14%. Notably, it achieved those cost savings — much of which it credited to the use of AI — while generating 23% higher revenues.
(The value of the AI cost savings is highlighted by substantial increases in other non-operating expenses. For instance, in the same period Klarna reported a 44% increase in losses from people not paying their loans back while funding costs rose 67%.)
Klarna trimmed its workforce down from 5000 to 3800 heading towards its (now stalled) IPO. It claims that nine out of 10 employees already use AI in their daily work, but is aiming to hire over 100 engineers by 2025 in Poland. Siemiatkowski says that the new Warsaw hub will play a pivotal role in the company’s broader strategy to lead AI adoption. Overall then Klarna was able to use AI to help to manage processing and servicing costs, which rose more slowly than revenue.
Does AI automatically mean returns for fintech? That is a more difficult question to answer. Shopify’s stock price is now around half what is was before the layoffs started. One explanation might be that while overall employee numbers are down, the average salary may rise as the hiring of expensive AI engineers continues. Klarna’s technology and product development expenses climbed 17% over the period discussed above, although the figures do show that Klarna’s cost savings more than offset the higher technology costs. Overall operating expenses fell 2% and the company posted a small profit on a net income basis in the third quarter last year.
xxx
Called Monzo Stand-in, the replica platform enables customers to make payments, withdraw cash, freeze their card and send and receive bank transfers even if the main Monzo app goes down.
Outages in the banking industry have become a pressing concern for regulators and politicians as more customers move online.
Recent figures from the UK Government Treasury Committee revealed that nine of the top banks and building societies operating in the UK accumulated at least 803 hours – the equivalent of more than 33 days – of tech outages in the last two years.
During a recent three-day outage at Barclays, 56% of online payments failed due to ‘severe degradation’ of its mainframe processing performance. The bank confirms it expects to pay between £5 million and £7.5 million in compensation to customers for ‘inconvenience or distress’.
In February British banks TSB, Bank of Scotland, Nationwide Building Society, Halifax, Lloyds and First Direct all reported problems for ingoing and outgoing payments.
Monzo Stand-in is an independent set of systems that run on Google Cloud Platform and is able to take over from the bank’s Primary Platform, which runs in Amazon Web Services (AWS), in the event of a major incident. It supports the most important features of Monzo like spending on cards, withdrawing cash, sending and receiving bank transfers, checking account balances and transactions, and freezing or unfreezing cards.
xxx
Snippings