How Anthropic has doubled down on AI safety

Anthropic, which was formed by a group of safety-conscious defectors from OpenAI, spent 2023 establishing itself as one of just a handful of envelope-pushing developers of commercial AI models.

The company raised almost $7 billion in 2023 (from Amazon and Google, among others), and last October it told investors that it was generating revenue at a $100 million annualized run rate. In July, the startup launched the second major release of its LLM, dubbed Claude 2, with improved performance on standardized tests, a larger memory for text input, and improved computer coding ability.

Numerous AI app developers use Claude 2’s API (often alongside those of other state-of-the-art models) for its unique skills in text summary and generation. On the safety side, the company implemented a novel way of keeping Claude away from toxic or dangerous content—an approach the company calls “Constitutional AI.”

Instead of laboriously providing a model with human feedback, the researchers give a model a list of general principles (a “constitution”) to follow when creating its responses. A second AI then monitors the first, continually critiquing how well it follows the principles.

Explore the full 2024 list of Fast Company’s Most Innovative Companies, 606 organizations that are reshaping industries and culture. We’ve selected the firms making the biggest impact across 58 categories, including advertising, artificial intelligence, design, sustainability, and more.

https://www.fastcompany.com/91033423/anthropic-most-innovative-companies-2024?partner=rss&utm_source=rss&utm_medium=feed&utm_campaign=rss+fastcompany&utm_content=rss

Creată 11mo | 19 mar. 2024, 12:20:03


Autentifică-te pentru a adăuga comentarii

Alte posturi din acest grup

This app combines Wikipedia and TikTok to fight doomscrolling

“Insane project idea: all of wikipedia on a single, scrollable page,” Patina Systems founder Tyler Angert posted on X earli

10 feb. 2025, 22:30:04 | Fast company - tech
Elon Musk’s $97 billion OpenAI bid would give the DOGE chief even more power in the AI race

A group of investors led by Elon Musk has given OpenAI an unsolicited offer of $97.4 billion to buy the nonprofit part of OpenAI. An attorney for the group submitted the bid to OpenAI Monday, the

10 feb. 2025, 22:30:04 | Fast company - tech
What exactly is the point of the AI Action Summit?

The world’s leading minds in AI are gathering in Paris for the AI Action Summit, which kick

10 feb. 2025, 20:10:10 | Fast company - tech
NASA astronauts are streaming live on Twitch  from space. Here’s how to watch

Ever wondered what life is like for an astronaut? Now you can ask during NASA’s first

10 feb. 2025, 20:10:08 | Fast company - tech
Credo AI’s vision for ethical and transparent AI governance

Brendan Vaughn, editor-in-chief of ‘Fast Company,’ interviews Credo AI’s CEO on AI governance trends at the World Economic Forum 2025.

https://www.fastcompany.com/91275783/credo-ais-vision-fo

10 feb. 2025, 17:50:05 | Fast company - tech
AI summit in Paris brings together Big Tech and U.S. VP Vance

Major world leaders are meeting for an AI summit in Paris, where challenging diplomatic talks are expected as tech titans fight for dominance in the

10 feb. 2025, 17:50:03 | Fast company - tech
Roblox joins $27 million industry nonprofit to support online safety

A group of internet businesses, including Roblox, Google, OpenAI, and Discord, have cofounded a nonprofit called Robust Open Online Safety Tools (ROOST).

The new organization will fund f

10 feb. 2025, 15:30:08 | Fast company - tech