Centar Municipality Raises Newborn Support to 10,000 Denars - 4 Million Budget for 2026
17.05.2026
17.05.2026
17.05.2026
17.05.2026
17.05.2026
17.05.2026
17.05.2026
17.05.2026
16.05.2026
15.05.2026
17.05.2026
17.05.2026
16.05.2026
17.05.2026
17.05.2026
17.05.2026
17.05.2026
16.05.2026
15.05.2026
17.05.2026
17.05.2026
17.05.2026
09.03.2026
27.02.2026
19.02.2026
14.04.2026
07.11.2025
07.11.2025
No news available in this category.
23.04.2026
23.04.2026
12.04.2026
OpenAI has opened the door to a lot more than recipe and essay questions. From Friday, ChatGPT Pro users in the United States can link their bank accounts to the chatbot and ask it about everything - from spending analysis to retirement planning. Two years ago, you wouldn't have given this data to a banker. Today, you're handing it to a language model.
The link runs through Plaid - the financial middleware already used by Robinhood, Venmo and hundreds of other apps. Through it, ChatGPT gets access to more than 12,000 financial institutions, including Schwab, Fidelity, Chase, American Express and Capital One. Once connected - a dashboard with portfolio performance, subscriptions, upcoming payments.
The timeline is interesting. A month ago, OpenAI bought the team from the startup Hiro, backed by Ribbit, General Catalyst and Restive. According to the company, their expertise was key to launching this product - though it doesn't say whether the team built the whole thing. An integration with Intuit is also planned, which will allow analysis of the tax implications of selling stock or the odds of being approved for a credit card.
The number OpenAI is highlighting: more than 200 million users are already asking ChatGPT financial questions every month. That isn't a research finding - it is an admission that people are already going to a language model for financial advice, and the company is now simply formalising the relationship. The new GPT-5.5 model, they say, is stronger at reasoning with context - precisely what you need for calculations and money analysis.
The appeal is clear, the risks too. This data is perhaps the most sensitive that exists - more sensitive than medical data, in certain ways. OpenAI says the data will be deleted 30 days after an account is disconnected. What happens between day one and day 30, and where all of it ends up - that is what lawyers will be cutting into thin slices for years to come. Perplexity and Anthropic have already entered the same space. The question isn't whether this is a trend - it is how long until the first big incident.
The latest 10 news from this category
Kevin O'Leary's Stratos will raise daytime temperatures by 2.8 degrees and night-time by 15.5 - and the Croatian Pantheon, nine...
Sam Altman is unhappy with how ChatGPT is buried in iOS - Apple, for its part, has its own complaints...
Nine years of hardware grind, one failed IPO, OpenAI and AWS as new customers - and suddenly Benchmark beats the...
For the first time, Musk's company published detailed descriptions of every traffic incident. The irony: some of them happened precisely...
Pro users in the US can connect their entire financial portfolio to ChatGPT through Plaid. What could go wrong with...
A new feature copied from Snapchat and BeReal is causing chaos - photos go to all friends with a single...
A venture capitalist with 18 months of experience nearly missed the 2016 meeting. Today his firm holds shares worth over...
A California jury gets three questions that could change the structure of the world\'s biggest AI company. Musk claims his...
No filters, no opening the gallery, no screenshots - the feature is launching globally. The question is whether Meta is...
The nine-year-old startup doubled its valuation in a year. A contract for a space missile shield, a contract with the...