Last updated: August 12th, 2026 at 11:53 UTC+02:00
SamMobile has affiliate and sponsored partnerships, we may earn a commission.
Anthropic's AI coding tool seems to be making mistakes too.
Reading time: 3 minutes
Abhijeet Mishra / SamMobile
Samsung logo
According to a new report from Chosun Biz, Samsung’s System LSI division has started using Anthropic’s AI coding tool, Claude Code, for semiconductor-specific tasks.
The tool has reportedly helped the brand significantly reduce development times in some cases. The publication provides two examples of how Claude Code has helped Samsung’s engineers.
First, a custom SoC verification task that would normally take more than a month was completed in about two days with the help of Claude Code. The task involved verifying internal data connections in a custom SoC, with Claude helping engineers build a virtual verification environment and test scenarios.
In another case, a second-year engineer used Claude Code to develop USB device models for an emulator and adapt an Android driver. The work, which would normally take around a month, was completed in just one day with the help of the AI tool.
Samsung first gave its software developers access to Claude Code in May 2026. Later, the company expanded its use to semiconductor development and verification work.
However, Claude Code has also made mistakes. When instructed to fix errors, the AI tool occasionally masked error messages instead of properly addressing the underlying issues. In some cases, it also reverted unrelated tasks that had already been completed or attempted to modify core Register Transfer Level (RTL) circuit designs that it wasn't supposed to touch. These issues reportedly occurred because the AI lacked sufficient context about hardware dependencies.
As a result, Samsung is treating Claude Code as an assistant rather than allowing it to work as an autonomous engineer. Engineers still need to review its output to ensure that the AI doesn't introduce errors or make unintended changes to semiconductor designs, where even seemingly minor mistakes can have significant consequences.
The publication also points out that Samsung’s System LSI division has around 6,000 employees, whereas one of its major competitors, Qualcomm, has around 52,000 employees. The report suggests that AI could help Samsung offset the disadvantage of having a much smaller workforce while improving development efficiency.
The use of Claude Code is also part of Samsung’s broader push to integrate generative AI across its businesses. The company has expanded its AI toolkit to include other services such as Google Gemini and ChatGPT, with its broader AI transformation efforts covering areas including R&D, manufacturing, marketing, and support.
First Samsung device: Galaxy S II
I’m a computer science engineer living in Hyderabad, India, who has a keen interest in automobiles and consumer electronics. My journalism career kicked off in 2017 with MySmartPrice where I wrote news, features, buying guides, and explanatory articles about technology among other things, and reviewed many products, including smartphones, tablets, laptops, PC components, smartwatches, audio devices, wearables, and smart home products. Since then, I have worked for 91Mobiles, Apple, and Onsitego, before finally landing on SamMobile.