AI Agent 'Lobster Fever' Grips China Despite Risks

A man wears a lobster hat that represents the OpenClaw logo, an open-source AI assistant at the Baidu headquarters in Beijing on March 11, 2026. (Photo by ADEK BERRY / AFP)
A man wears a lobster hat that represents the OpenClaw logo, an open-source AI assistant at the Baidu headquarters in Beijing on March 11, 2026. (Photo by ADEK BERRY / AFP)
TT

AI Agent 'Lobster Fever' Grips China Despite Risks

A man wears a lobster hat that represents the OpenClaw logo, an open-source AI assistant at the Baidu headquarters in Beijing on March 11, 2026. (Photo by ADEK BERRY / AFP)
A man wears a lobster hat that represents the OpenClaw logo, an open-source AI assistant at the Baidu headquarters in Beijing on March 11, 2026. (Photo by ADEK BERRY / AFP)

Chinese entrepreneur Frank Gao used to spend long hours running his social media accounts but now outsources the chore to AI agent tool OpenClaw, which is taking the country by storm despite official warnings over cybersecurity.

OpenClaw, created in November by an Austrian coder, differs from bots like ChatGPT because it can execute real-life tasks such as sending emails, organizing files or even booking flight tickets.

"Since January, I've spent hours on the lobster every day," Gao told AFP, referring to OpenClaw's red crustacean mascot. "We're family."

After downloading OpenClaw, users connect it to existing artificial intelligence models of their choice, then give it simple instructions through instant messaging apps, as if to a friend or colleague.

The tool has fascinated tech circles worldwide but particularly in China, gripping tech-savvy companies and individuals keen to keep up with the next big thing in AI.

Hundreds of people queued at tech giant Baidu's Beijing headquarters this week for an OpenClaw event where engineers helped attendees set up their "little lobsters".

It was one of many similar meetups to experiment with the tool, which are drawing crowds from Shanghai to Shenzhen.

Some municipalities, including the eastern cities of Wuxi and Hangzhou, have pledged hundreds of thousands of dollars to support the adoption and development of OpenClaw and other AI agents.

But the lobster fever, as it has been dubbed, has also sparked security concerns.

"What's truly scary about agents like OpenClaw is this: once they have your digital keys, they can theoretically access all the services you've authorized, and can autonomously decide when to activate them," Gao warned.

"The attacker effectively gains a 'master key' to your digital identity," said the engineer, who has named his OpenClaw agent "Q" after his business name QLab.

- 'Use with caution' -

Chinese national cybersecurity authorities and Beijing's ministry of industry and IT have warned of the risks of OpenClaw hacks.

"Use intelligent agents such as 'lobster' with caution," national IT research institute expert Wei Liang advised government agencies, public institutions, companies and individuals in a message on state media.

The mixed signals of rolling out policy incentives while issuing warnings "reflects the authorities' cautious tolerance towards 'lobster fever'," Zhang Yi, founder of tech consultancy iiMedia, told AFP.

Austrian programmer Peter Steinberger, who built OpenClaw to help organize his digital life, was hired last month by ChatGPT maker OpenAI.

Meanwhile, a separate team of coders that made Moltbook, a Reddit-like pseudo social network where OpenClaw agents converse, are joining Meta.

Top Chinese tech companies have also been quick to get involved.

The likes of Tencent, Alibaba, ByteDance and Baidu are offering simplified installation and affordable coding plans to help users who want to host OpenClaw agents on their cloud servers -- seen as safer than downloading it onto a personal computer.

In recent days AI companies big and small have also launched their own competing agent tools, such as ByteDance's ArkClaw, Tencent's WorkBuddy and Zhipu AI's AutoClaw.

The relatively low cost for cloud deployment of OpenClaw in China, subsidised by big tech firms, is one factor behind its popularity, said Gao Rui, a senior product manager at Baidu AI Cloud.

"For most people, it's likely just the price of a cup of coffee... which is why people will probably be keen to give it a try," she told AFP.

- FOMO -

Fear of missing out is also a big driver behind OpenClaw's success in China, said Chen Yunfei, an AI developer who created a popular online guide for using the tool.

"Most Chinese people are quite studious and forward-looking, so when confronted with new things, they might have stronger feelings" of so-called FOMO, he said.

Xie Manrui, a programmer whose latest project is a visualized system for managing OpenClaw agents, said the tool had arrived "at the right moment" to change perceptions in China of what AI can do.

"For many, AI is merely a clever chatbot that talks all the time but cannot act," he said.

Either way, it has piqued the curiosity of many young users.

At the Baidu event in Beijing, 24-year-old college student Zheng Huimin was waiting patiently in line with her friends.

"I'd like to give it a go to see what tasks it can actually help me accomplish," she told AFP.



US House Speaker Warns Curbing AI Too Sharply Could Leave China Ahead

US Speaker of the House Mike Johnson speaks during the second day of the Republican National Committee midterm convention at the American Airlines Center in Dallas, Texas on September 10, 2026. (AFP)
US Speaker of the House Mike Johnson speaks during the second day of the Republican National Committee midterm convention at the American Airlines Center in Dallas, Texas on September 10, 2026. (AFP)
TT

US House Speaker Warns Curbing AI Too Sharply Could Leave China Ahead

US Speaker of the House Mike Johnson speaks during the second day of the Republican National Committee midterm convention at the American Airlines Center in Dallas, Texas on September 10, 2026. (AFP)
US Speaker of the House Mike Johnson speaks during the second day of the Republican National Committee midterm convention at the American Airlines Center in Dallas, Texas on September 10, 2026. (AFP)

House Speaker Mike Johnson on Sunday warned that moving too quickly to curb AI development could cost the United States the tech race to China.

Johnson's comments come amid a growing chorus of AI leaders and experts sounding the alarm that untamed AI growth could cause irreparable damage to humanity.

"If Congress just races in and does some sort of emergency session to try to regulate AI, we will lose the race to China, and that is a threat to every single American," Johnson said on CNN Sunday morning.

"But we don't need everybody to panic right now. We need to handle this new technology like we have others in the past, and make sure we're doing everything we can responsibly to also not smother American innovation," he added. "We have to do both things simultaneously."

Johnson, a Republican, called for a meeting in Congress with top AI executives to figure out how best to regulate the industry. He added that he had discussed the idea with President Donald Trump, and, when asked if such a meeting could take place this week, Johnson said: "Yeah, I'd do it tomorrow."

The remarks follow a call by Anthropic CEO Dario Amodei to slow down the development of the powerful technology, amid mounting worries over the risks of "superintelligent" computer systems.

His comments came a few days after AI researcher Jacob Coxon, who had previously worked at both OpenAI and Anthropic, decided to leave the industry, accusing both US companies of "gambling with our lives" in the race to develop models capable of self-improvement.

OpenAI boss Sam Altman and Elon Musk, who owns xAI, quickly chimed in to say they agreed with Amodei's assessment, as pressure builds for improved oversight.

Concerns have grown in recent weeks after OpenAI revealed that during testing, AI models broke out of their confined environment, connected to the internet and infiltrated Hugging Face, a site developers use to store and share code.

In that incident and others, the blame was placed on AI agents, which are software programs that can carry out tasks without constant supervision by humans.

In early August, the US government put in place a voluntary security review process for advanced AI models before their release, but the parameters of that program remain unclear.

Industry observers have questioned how effective it would be, as the Trump administration has thus far favored a light-touch, deregulatory approach to most industries, including tech.


Anthropic Boss Calls for Slowing Pace of AI Development

(FILES) Anthropic CEO Dario Amodei looks on as he takes part in a session on AI during the World Economic Forum (WEF) annual meeting in Davos on January 23, 2025. (Photo by FABRICE COFFRINI / AFP)
(FILES) Anthropic CEO Dario Amodei looks on as he takes part in a session on AI during the World Economic Forum (WEF) annual meeting in Davos on January 23, 2025. (Photo by FABRICE COFFRINI / AFP)
TT

Anthropic Boss Calls for Slowing Pace of AI Development

(FILES) Anthropic CEO Dario Amodei looks on as he takes part in a session on AI during the World Economic Forum (WEF) annual meeting in Davos on January 23, 2025. (Photo by FABRICE COFFRINI / AFP)
(FILES) Anthropic CEO Dario Amodei looks on as he takes part in a session on AI during the World Economic Forum (WEF) annual meeting in Davos on January 23, 2025. (Photo by FABRICE COFFRINI / AFP)

Anthropic CEO Dario Amodei on Saturday called on artificial intelligence (AI) firms to slow down the development of the powerful technology, amid mounting worries over the risks of "superintelligent" computer systems.

His comments came a few days after AI researcher Jacob Coxon, who left OpenAI to join Anthropic, decided to leave the industry, accusing both US companies of "gambling with our lives" in the race to develop models capable of self-improvement.

"We must slow the pace at which we improve the capabilities of AI models. Progress will still seem fast, and we must make wise use of the time we gain," Amodei wrote in a post on his personal website.

"AI brings risks, and because it is such a powerful technology, these risks are serious."

"Not building the technology deprives humanity of benefits or simply places AI in the hands of authoritarian powers, while building it too fast is reckless. We have sought a middle way," said Amodei, who co-founded Anthropic with his sister Daniela in 2021.

"But over the last few months, I have become convinced that fully addressing the risks requires even more prudence."

According to AFP, Amodei's statement echoes another call by Anthropic in early June to slow or suspend development.

Then in late July, Sam Altman, the boss of Anthropic rival OpenAI, said developers might need to voluntarily pump the brakes on their rapid advances to give society a chance to catch up.

At roughly the same time, more than 1,000 employees at cutting-edge AI companies including Amodei signed a petition calling on the US government to help "deliberately pace the frontier of automated AI development."

OpenAI had revealed that during testing, the AI models broke out of their confined environment, connected to the internet and infiltrated Hugging Face, a site developers use to store and share code.

Earlier this week, Coxon, 27, sounded the alarm after spending the past three years pretraining AI models -- first at OpenAI and then, this year, at its rival Anthropic, which he considered more cautious in its approach.

Pretraining is the stage where AI models absorb vast quantities of data.

"Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives," Coxon said Tuesday.

"The people building AI earnestly believe that it could kill us all by the end of the decade," he said in a post on X.
Superintelligence is the theoretical point when AI's capabilities exceed human intelligence.


Oracle Shares Rise as AI Cloud Backlog Beats Estimates

FILE PHOTO: A logo of cloud service provider Oracle is seen at the company's offices at Eastpoint Business Park, Dublin, Ireland October 18, 2021. REUTERS/Tom Bergin/File Photo
FILE PHOTO: A logo of cloud service provider Oracle is seen at the company's offices at Eastpoint Business Park, Dublin, Ireland October 18, 2021. REUTERS/Tom Bergin/File Photo
TT

Oracle Shares Rise as AI Cloud Backlog Beats Estimates

FILE PHOTO: A logo of cloud service provider Oracle is seen at the company's offices at Eastpoint Business Park, Dublin, Ireland October 18, 2021. REUTERS/Tom Bergin/File Photo
FILE PHOTO: A logo of cloud service provider Oracle is seen at the company's offices at Eastpoint Business Park, Dublin, Ireland October 18, 2021. REUTERS/Tom Bergin/File Photo

Oracle shares rose 5.5% premarket on Friday after the cloud computing and software company's stronger-than-expected quarterly results eased concerns around its massive debt-driven spending spree.

The Austin, Texas-based firm booked more than $30 billion of additional AI cloud contracts in the first fiscal quarter, boosting its revenue backlog to $664 billion, above analyst estimates of $639.89 billion, according to data from Visible Alpha.

Oracle's upbeat results follow a period of underperformance, as the company races ⁠to keep pace ⁠with hyperscale rivals, with mounting debt and squeezed cash flows raising doubts over when its massive AI spending will pay off.

The results should address key investor concerns including whether Oracle's backlog growth can be ⁠sustained, its conversion into revenue amid data center delays, and the need for further capital raises, Reuters quoted J.P. Morgan analysts as saying.

Its shares were down more than 21% so far this year, compared with a nearly 11% rise in the benchmark S&P 500 index.

"Oracle's problem has not been finding customers, but proving that its enormous data center build-out can eventually generate ⁠enough ⁠cash to justify the cost. The results should nevertheless ease fears that Oracle is building ahead of demand," said Lale Akoner, etoro market strategist.

Oracle is set to add about $24 billion in market value at the current share price of $161.3, if gains hold.

The stock trades at 16.86 times its forward earnings estimates, compared with Microsoft's 23.84 multiple and Amazon's 22.58, according to data compiled by LSEG.