Anthropic Unveils AI Model Capable of Working Nearly an Entire Day Without Human Input

Artificial intelligence startup Anthropic has introduced a powerful new model that can operate autonomously for close to seven hours – a development that signals a potential shift in how businesses integrate AI into daily operations.

The new model, Claude Opus 4, was launched on Thursday alongside a general-purpose version called Claude Sonnet 4. According to Anthropic, the Opus model is designed to tackle complex, multi-step tasks over extended periods, marking a step toward AI that can handle broader objectives, much like a human employee would.

Unlike earlier AI systems that excelled at isolated tasks, Claude Opus 4 can maintain focus and deliver results across a full-length project without needing frequent human prompts. While it doesn’t yet replicate the full scope of a typical workday – such as multitasking or spontaneous collaboration – it represents a leap in long-form reasoning and autonomy.

“It’s the kind of work that might take up 30% of your day – things that are necessary but not particularly rewarding,” said Scott White, Product Lead at Anthropic. He noted the assistant is meant to augment professionals by automating routine or time-consuming parts of their jobs.

As an example, White described how a marketing professional could rely on Claude Opus 4 to review past campaign data across platforms like Facebook and Google, identify performance trends, and generate actionable insights for a future strategy – without manual intervention at each step.

“It’s basically Claude being able to think through your objective over a long stretch of time, using reasoning and tools to keep pushing forward,” White said.

Anthropic’s release arrives amid a surge in enterprise investment in generative AI. A recent survey by Menlo Ventures, an Anthropic backer, reported a sixfold increase in corporate AI spending in 2024 compared to the previous year. The study also suggested Anthropic is gaining traction, gradually narrowing the gap with industry leader OpenAI.

Meanwhile, tech heavyweights are deepening their own AI initiatives. Microsoft just announced a new coding assistant for GitHub, Google unveiled its autonomous developer tool “Jules,” and Apple is reportedly exploring a coding assistant with help from Anthropic.

But as AI gains capabilities, concerns over its impact on employment continue to grow. The World Economic Forum reported earlier this year that over 40% of employers anticipate reducing staff as AI becomes more embedded in their operations. LinkedIn’s Aneesh Raman recently warned that entry-level roles may be especially vulnerable.

White acknowledged these concerns and called for broader dialogue. “This isn’t a challenge Anthropic can solve alone,” he said. “Governments, policymakers, and businesses all need to collaborate to responsibly guide how AI is adopted and managed.”