An Anthropic report on Claude misuse, Intel restructuring and layoffs: This week’s news round-up

The Washington Post gets cosy with OpenAI, Anthropic underlines its commitment to safety and Intel proposes more layoffs and restructuring, here are the news stories that caught our attention this week.

Washington Post partners with OpenAI

The Washington Post, owned by Amazon founder Jeff Bezos, has partnered with OpenAI to integrate the newspaper's journalism into ChatGPT. ChatGPT now provides users with summaries, quotes, and direct links to original articles from The Washington Post. The newspaper said it continues to be LLM-agnostic and build its own range of AI-powered solutions for both its business and its users.

Synthesia reaches $100m ARR and readies AI agent

UK AI video generation startup Synthesia is set to launch a new AI agent tool designed to make videos more interactive, according to Sifted. The AI agent will be released in “the first half of this year” and is aimed at improving user engagement by allowing viewers to interact with video content, potentially transforming passive viewing into an active experience. Synthesia also reached over $100 million in annual recurring revenue (ARR) in April and received a strategic investment from Adobe Ventures. This follows a $180 million funding round in January.

Intel to lay off a fifth of workforce

Chip maker Intel plans to lay off over 20% of its workforce in a bid to streamline operations and reduce bureaucratic inefficiencies. As Intel reported its Q1 earnings CEO Lip Bu-Tan also sent a message to employees that lays out plans for restructuring and expands the company’s return to office mandate from three days to four. The layoffs are thought to affect about 22,000 staff, based on Intel's global workforce of 108,900 at the end of 2024.

Anthropic underlines commitment to safety with report on misuse of Claude AI

In a bid to assure us of its commitment to transparency and a proactive culture of safety, Anthropic has published a report showing how its Claude AI model has been exploited by malicious actors and which details the company’s efforts to mitigate this misuse.

The report, called Detecting and Countering Malicious Uses of Claude: March 2025, reports that, for instance:

• An influence-as-a-service operation used Claude.to orchestrate over a hundred social media bot accounts to clients’ political narratives.
• Claude was leveraged in a recruitment fraud campaign targeting job seekers in Eastern European countries. 
• A novice actor with limited technical skills leveraged Claude to develop malware that would typically require more advanced expertise.

According to Anthropic: “This report outlines several case studies on how actors have misused our models, as well as the steps we have taken to detect and counter such misuse.”

The report said that users are starting to use AI frontier models to semi-autonomously orchestrate complex abuse systems that involve many social media bots. The company expects this trend to continue as agentic AI systems improve. It also points out that generative AI can accelerate capability development for less sophisticated actors, potentially allowing them to operate at a level previously only achievable by more technically proficient individuals.

Anthropic has been largely commended for its transparency in publishing this report. On LinkedIn for example, commentators praised the company for its commitment to safety.

More from Mind the Product

Anthropic secures further funding to compete against OpenAI

DeepSeek AI launch: The view from product leaders

AI and ML in e-commerce. Strategies, trends, and future insights