EUR/USD
1.14859
0.096%
Gold
4376.53
0.805%
Oil
95.469
-1.085%
USD/JPY
156.726
0.486%
GBP/USD
1.33935
0.269%
GBP/JPY
209.919
0.759%
Real-time News
On September 19th, Reuters reported that Anthropic is considering launching a new AI model to counter the momentum gained by OpenAI since the release of GPT-6Astra, ahead of its upcoming IPO and following its CEOs call for an industry-wide slowdown. OpenAIs GPT-6Astra, released this month, has garnered significant attention in the enterprise market, raising concerns among some investors. They told Reuters they are reassessing Anthropics position as a leading provider of enterprise-grade AI tools. One source said Anthropic is evaluating the security of its next-generation model as one consideration before releasing it. Sources familiar with the companys thinking said some discussions involve balancing investment in a new model release with improving the companys profitability, as rising interest rates have made investors more focused on the timing of expected profit realization. Two other sources indicated that Anthropic may postpone its IPO until after the US midterm elections in November, with the elections not expected to significantly impact the offering. The IPO has already been delayed, with Reuters previously reporting that the roadshow was expected to begin as early as mid-October.According to Reuters, Anthropic is considering launching a new artificial intelligence model before its IPO; Anthropics initial public offering may be delayed until after the U.S. midterm elections.On September 19th, Houthi spokesman Yahya Sarreya issued a statement on the 18th claiming that Saudi Arabian warplanes had launched 26 airstrikes against Yemens Taiz province in the past 24 hours. This week, Saudi warplanes have launched 300 airstrikes against Yemen. Saudi Arabia has not yet responded to the Houthi claims.On September 19th, Google announced that its Gemini AI model accidentally accessed the internet and entered the systems of three companies during a cybersecurity capability test. This is the first known instance of a Google AI system autonomously performing such an operation. The incident occurred in May and was conducted by security firm Irregular. During the test, the Gemini model was performing a task in a "capture the flag" cybersecurity exercise. However, because the virtual company names in the test environment were identical to those of real companies, and the model unexpectedly gained internet access, it entered the real enterprise systems. Google stated that in one instance, the model entered a protected system by trying passwords; in two other instances, it discovered credentials in a public network repository and attempted to access the relevant systems. Once the model confirmed access to the real company systems, it stopped further operations. Google stated that the incident did not cause any damage to the companies involved and therefore was not considered a case of model malfunction, adding that security mechanisms helped the model stop its behavior in time. The affected companies have been notified, and Google has also reported the incident to relevant regulatory agencies. This incident is similar to previous security test incidents disclosed by AI companies such as OpenAI and Anthropic, once again drawing attention to the ability of AI models to autonomously perform cyber operations and the mechanisms of security protection.Ukrainian President Volodymyr Zelensky thanked US President Donald Trump for signing the Russian sanctions bill and thanked members of Congress for their support.
