ME News, June 26 (UTC+8), reveals internal divisions within the company regarding governance evaluation. Some former employees disclosed that there is an active debate atmosphere within the company; however, other former employees painted a more somber picture, stating that sharp criticism is largely confined to private group chats, and regular all-hands meetings are even jokingly referred to by employees as "Dario Vision Quests." Listening to CEO Dario Amodei speak is like "listening to a pastor's sermon (sermon)," and few people challenge management decisions in public meetings. The root of this somber atmosphere lies in Anthropic's deeply ingrained elite security belief of "attack as defense." Senior executives internally use OpenAI CEO Sam Altman (as well as Meta and xAI) as negative examples, while instilling elite logic in employees: only by maintaining industry leadership in business, computing power, and research talent can one have a say in rule-making. Former OpenAI board member Helen Toner once made a forest analogy: villagers will inevitably flood into a magic forest full of monsters; Anthropic's strategy is to venture deep into the forest first, taming the monsters while releasing technological dividends and controlling catastrophic risks. The pursuit of technological dominance has also repeatedly put Anthropic's safety and ethical choices under controversy. In the fall of 2024, management, despite internal opposition, partnered with Palantir to open services to U.S. intelligence and defense agencies. Currently, Claude has been confirmed to be used for target identification in Middle East conflicts. When questioned about attacks causing civilian casualties, Amodei stated he was unaware but argued that "as long as humans make the final decision, the use of technology is compliant." In June this year, the development team secretly embedded sabotage code in the flagship model Claude Fable 5 to covertly interfere with non-compliant development. However, after encountering strong industry backlash, the company was forced to compromise, announcing that the covert interference mechanism would be transformed into publicly visible security restrictions. Industry critics point out that betting frontier AI safety solely on the moral superiority of a few elites inevitably leads to unavoidable blind spots in self-discipline. (Source: BlockBeats)

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

Reward
like
Comment
Repost
Share

Comment

Add a comment

No comments

Trending Topics
View More
#
Get2SharesOfSKHynixAtZeroCost
1.54M Popularity
#
MicronOvertakesMetaInMarketValue
167.92K Popularity
#
WorldCup🇫🇷vs🇳🇴
322.15K Popularity
#
USMayPCEInflationRisesTo4.1%HighestIn3Years
550.05K Popularity
#
StakeUSD1Earn9.48%APR
981.19K Popularity

Pinned

Sitemap

Under the banner of preventing Altman from going wrong, power was centralized—and former employees have revealed that Anthropic is instead deeply trapped in the “good people” myth, with a one-voice, top-down rule prevailing.

Trending Topics

Get2SharesOfSKHynixAtZeroCost

MicronOvertakesMetaInMarketValue

WorldCup🇫🇷vs🇳🇴

USMayPCEInflationRisesTo4.1%HighestIn3Years

StakeUSD1Earn9.48%APR

Pinned