cd /news/artificial-intelligence/anthropic-launches-opus-5-5-its-firs… · home topics artificial-intelligence article
[ARTICLE · art-137294] src=ca.finance.yahoo.com ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

Anthropic launches Opus 5.5, its first model since CEO Amodei called for AI slowdown

Anthropic debuted Opus 5.5 on Tuesday, the first model in its Claude 5.5 family and the first release since CEO Dario Amodei called for a slowdown in frontier AI development. Anthropic said Opus 5.5 performs on par with its Fable 5.1 model while costing 40% less than Opus 5, at $4 per million input tokens and $20 per million output tokens versus $5 and $25 for Opus 5. The company said the Claude 5.5 models underwent extensive alignment testing, external evaluations, and safeguards for high-risk areas including cybersecurity and biology.

by read2 min views1 publishedSep 22, 2026
Anthropic launches Opus 5.5, its first model since CEO Amodei called for AI slowdown
Image: Ca (auto-discovered)

Anthropic (ANTH.PVT) on Tuesday debuted its latest high-powered AI model, Opus 5.5, the company's first in its Claude 5.5 family. It's also the first model since CEO Dario Amodei called for a slowdown in developing frontier AI.

According to Anthropic, the new model offers performance on par with its Fable 5.1 model but costs 40% less than Opus 5.

Pricing of AI models has become a key competitive advantage for AI companies. Anthropic said Opus 5.5 will cost $4 per million input tokens and $20 per million output tokens. The company's Opus 5 costs $5 per million input tokens and $25 per million output tokens.

Tokens are units of measurement for AI models that represent small words or word fragments.

As for safety and security, Anthropic said its Claude 5.5 models have undergone extensive alignment testing, external evaluations, and "safeguards for high-risk areas like cybersecurity and biology."

The company also said it is looking to improve rewards for AI alignment and is working on developing "new safety training scenarios."

The AI industry is dealing with fears that the technology could go out of control and cause widespread harm.

Earlier this month, Anthropic researcher Jacob Coxon posted on X that he was resigning from his position, saying that neither the company nor his former employer, OpenAI (OPAI.PVT), took AI safety seriously. He also said that the people building the technology "earnestly believe that it could kill us all by the end of the decade."

Anthropic alignment science lead Evan Hubinger replied to Coxon's post, saying the former Anthropic worker is correct and that he believes there is a greater than 10% chance that the technology could "kill all humans."

Shortly thereafter, Amodei posted an essay calling for a slowdown in AI development. OpenAI CEO Sam Altman and SpaceXAI's (SPCX) Elon Musk said they agreed with Amodei's statement.

But others in the industry, such as Nvidia (NVDA) CEO Jensen Huang, pushed back on the fears that AI will destroy us all, saying such concerns aren't grounded in science.

President Trump has also derided AI fears as a conspiracy.

Email Daniel Howley at dhowley@yahoofinance.com. Follow him on Twitter at @DanielHowley*.*

Read the latest financial and business news from Yahoo Finance.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @anthropic 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/anthropic-launches-o…] indexed:0 read:2min 2026-09-22 ·