Katarzyna Petru
As noted by Benj Edwards from arstechnica, firms dealing with artificial intelligence are increasingly shifting the direction of their product development. Instead of a single chatbot, the user now manages a team of AI agents that share tasks and work in parallel. This approach is promoted by among others. Anthropic and OpenAI in their latest tools. In practice, this means that the role of humans is changing; instead of writing and analysing independently, the user supervises the work of AI, corrects mistakes, and decides what to do next. Although this sounds futuristic, current agents still require constant oversight and are not yet fully independent "workers". Nevertheless, the industry is heavily investing in this model, seeing it as the future of working with AI. Will managing bots actually prove to be more effective than traditional human work?
Claude gains a new, more powerful model
Anthropic has presented Opus 4.6, a major update to its flagship AI model and successor to version 4.5. For the first time in the Opus series, the model supports a context of up to 1 million tokens (in beta), allowing for the analysis of very long texts and complex code snippets in a single session. According to the company, the new version performs better than earlier competitor models in programming, reasoning, and information retrieval tests, although the latest OpenAI model has regained an advantage in one of the benchmarks. Opus 4.6 has also significantly improved its ability to solve problems that are simple for humans but difficult for AI, and performs better with very long materials — which is critical when using teams of AI agents. API pricing remains unchanged from the previous version, and the model is already available in the Claude service, via API, and on major cloud platforms.
Market Turmoil
The launches of new AI-based tools coincided with an exceptionally fraught period in the financial markets. After Anthropic expanded its agent solutions with additional professional features, investors began to worry that AI companies would be able to take over tasks traditionally performed by classic SaaS services. The market's reaction was swift and severe — hundreds of billions of dollars vanished from the valuations of companies in the software and financial services sectors within a matter of days. The most significant declines affected companies whose business models may be particularly susceptible to AI-based automation, and the wave of sell-offs extended beyond the American market. An additional source of concern emerged from OpenAI's announcements about developing a platform that allows AI agents to execute tasks independently within corporate systems. Although producers emphasise that this is not about replacing existing software, the market clearly perceives a shift in direction.
source: arstechnica
