Indonesian Political, Business & Finance News

Anthropic Says It Is Approaching AI That Can Build Itself

| Source: Antara (Indonesian) Translated from Indonesian | Technology
Anthropic Says It Is Approaching AI That Can Build Itself
Image: Antara (Indonesian)

Jakarta (ANTARA) — Artificial intelligence company Anthropic says it is getting closer to developing a model that can build itself, a process known as recursive self-improvement (RSI).

The news may have received little attention because Anthropic has just released its latest financial report, which shows that although the company’s revenue has risen, its expenses have also grown substantially.

According to TechCrunch on Monday (2 November), the RSI development was mentioned only as a small detail in the report.

The report says the company, led by Dario Amodei, is at a crossroads in the creation of AI models capable of improving themselves.

RSI could make AI highly proficient at training itself, triggering a dramatic acceleration in its capabilities.

It could also enable AI to build itself into a sufficiently powerful system to escape human control and potentially disregard the damage it causes while pursuing its goals.

For those with a more optimistic view of AI, however, RSI — or the use of reliable AI systems to test themselves and improve their performance — represents a vision of the technology as an unprecedented productivity tool.

Anthropic appears to believe that RSI has the potential to bring about either outcome.

In a financial filing submitted to the US government, Anthropic said it already uses AI in its research and development.

However, it warned: “Using AI to help build more capable AI can also create substantial safety risks, especially if it dramatically accelerates the pace of progress. This is one of the areas covered by our Responsible Scaling Policy, which sets out our plans to monitor and mitigate risks as we approach what is known as ‘recursive self-improvement’.”

The term RSI became popular in early 2026, although many initially suspected it was merely a passing trend.

First, the company has used AI to develop new models whose capability improvements it did not recognise at the time.

Second, it may soon decide on its own definition of RSI.

Anthropic says in the report that, while it remains at this crossroads, none of its models has crossed the RSI threshold. It has pledged to keep humans involved in the R&D process for the foreseeable future.

“We do not anticipate being able to use Claude safely to autonomously design and build generations of frontier models without meaningful human oversight or understanding, if and when that becomes technically possible. We are therefore considering updates to our Responsible Scaling Policy to ensure that we retain control over the development of frontier models at Anthropic,” the company said.

Sceptics may regard Anthropic’s report as a way to surprise the public while also serving as a marketing vehicle.

Even so, Anthropic’s progress with Claude, particularly in programming, mathematical problem-solving and cyber security, has been striking this year.

When the company says its AI is nearing RSI, it may be time for the public to prepare for the possibility of accelerated AI progress in the future.

More from 3 November 2026 →

Cite this article
Source: jawawa.id (https://jawawa.id/newsitem/anthropic-says-it-is-approaching-ai-that-can-build-itself-1793724775)
Originally published by Antara (Indonesian) (https://www.antaranews.com/berita/5830119/anthropic-makin-dekat-dengan-ai-mampu-membangun-dirinya-sendiri), translated from Indonesian
Print