Mistral launches a moderation API

Date:

Share post:


AI startup Mistral has launched a new API for content moderation.

The API, which is the same API that powers moderation in Mistral’s Le Chat chatbot platform, can be tailored to specific applications and safety standards, Mistral says. It’s powered by a fine-tuned model (Ministral 8B) trained to classify text in a range of languages, including English, French, and German, into one of nine categories: sexual, hate and discrimination, violence and threats, dangerous and criminal content, self-harm, health, financial, law, and personally identifiable information.

The moderation API can be applied to either raw or conversational text, Mistral says.

“Over the past few months, we’ve seen growing enthusiasm across the industry and research community for new AI-based moderation systems, which can help make moderation more scalable and robust across applications,” Mistral wrote in a blog post. “Our content moderation classifier leverages the most relevant policy categories for effective guardrails and introduces a pragmatic approach to model safety by addressing model-generated harms such as unqualified advice and PII.”

AI-powered moderation systems are useful in theory. But they’re also susceptible to the same biases and technical flaws that plague other AI systems.

For example, some models trained to detect toxicity see phrases in African American Vernacular English (AAVE), the informal grammar used by some Black Americans, as disproportionately “toxic.” Posts on social media about people with disabilities are also often flagged as more negative or toxic by commonly used public sentiment and toxicity detection models, studies have found.

Mistral claims that its moderation model is highly accurate — but also admits it’s a work in progress. Notably, the company didn’t compare its API’s performance to other popular moderation APIs, like Jigsaw’s Perspective API and OpenAI’s moderation API.

“We’re working with our customers to build and share scalable, lightweight, and customizable moderation tooling,” the company said, “and will continue to engage with the research community to contribute safety advancements to the broader field.”

Mistral also announced a batch API today. The company says it can reduce the cost of models served through its API by 25% by processing high-volume requests asynchronously. Anthropic, OpenAI, Google, and others also offer batching options for their AI APIs.



Source link

Lisa Holden
Lisa Holden
Lisa Holden is a news writer for LinkDaddy News. She writes health, sport, tech, and more. Some of her favorite topics include the latest trends in fitness and wellness, the best ways to use technology to improve your life, and the latest developments in medical research.

Recent posts

Related articles

Bluesky is courting the Swifties

Bluesky has grown by 2 million users — about 15% — since Donald Trump won the U.S....

Ford will pay up to $165M fine for rearview camera recall failures

Ford has agreed to pay a $165 million penalty to federal regulators after moving too slowly to...

Will Rivian be Volkswagen’s software savior? VW is betting $5.8B it will

Welcome back to TechCrunch Mobility — your central hub for news and insights on the future of...

ChatGPT can now read some of your Mac’s desktop apps

OpenAI’s ChatGPT is starting to work with other apps on your computer. On Thursday, the startup announced the...

New Apple security feature reboots iPhones after 3 days, researchers confirm

Apple’s new iPhone software comes with a novel security feature that reboots the phone if it’s not...

AI pioneer Francois Chollet leaves Google

Francois Chollet, a leading figure in the AI world, is leaving Google after close to a decade....

Amazon’s telehealth platform adds low-cost plans for hair loss, skin care, and more

Amazon One Medical is expanding its telehealth services with the launch of upfront and low-cost treatment plans...

Sales tax automation startup Kintsugi doubled its valuation this year

A 2018 Supreme Court ruling eliminated the requirement that an e-commerce retailer needed a physical location in...