Technology news and discussion on TalkFuse.
Here’s How Anthropic’s Latest Products Could Automate Your Job
Anthropic says its Claude Fable 5.2 model can complete 41.25 per cent of real freelance projects in a benchmark, a sharp rise from the 2.5 per cent achieved by the leading system when the test began in 2025. The result points to AI taking on a wider range of paid work, while the article highlights possible uses in design and visualisation.
Released on 9 November, Fable is the public version of Mythos 5.2, which is restricted to vetted researchers. The Remote Labor Index puts Claude Opus 5.5 and OpenAI’s GPT 6 Astra at about 21 per cent. Anthropic also announced Claude Sheets, Whiteboards and Data, for spreadsheets, digital whiteboards and data visualisation; an AI company founder separately demonstrated Fable turning 2D home plans into 3D walkthroughs.
In short
- Claude Fable 5.2 scored 41.25 per cent on a freelance-work benchmark.
- Anthropic released three apps for spreadsheets, whiteboards and data.
- A founder demonstrated turning floor plans into 3D walkthroughs.
Steelman: both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
The benchmark suggests AI is becoming capable of completing a wider range of paid projects, and tools for spreadsheets, visualisation and design could help people work faster or take on tasks that once required specialist skills. Supporters may see this as a way to make useful services more accessible and create new kinds of work, provided workers can adapt and share in the gains.
The case against
The same capabilities could let employers replace or reduce paid work, particularly for freelancers whose tasks can be completed to a specified brief. A benchmark result does not show how often businesses will adopt these tools or what the overall employment effects will be, but the rapid improvement may still justify concern about disrupted livelihoods and whether workers will have time and support to adjust.
Explained simply
Anthropic is an artificial intelligence company that makes Claude, a family of computer models designed to respond to instructions and carry out tasks. Such systems are increasingly being used for paid work, from writing and data analysis to design, though their ability to complete whole projects varies.
The Remote Labor Index tests how well AI systems handle real freelance assignments. Its results offer one way to compare models, but do not show how often businesses use them or whether they can replace a person across an entire job.
Anthropic’s tools include products for spreadsheets, digital whiteboards and data visualisation, alongside its Claude models. Their development matters to workers and employers because AI could change which tasks people do and how much work is done by software.