Each September, a who’s who of executives from across Silicon Valley descends on San Francisco’s Palace Hotel to charm investors at a conference hosted by the investment bank Goldman Sachs.

This past week, between talk of growth and potential returns, tech titans found themselves addressing the abrupt resignation of Anthropic researcher Jacob Coxon.

Coxon, a 27-year-old who worked at OpenAI before joining its chief rival Anthropic, said on Tuesday that people building artificial intelligence (AI) believed the technology could destroy humanity.

They are “gambling with our lives”, he said, “these will soon be superhuman systems that can hack anything”.

Coxon is by no means the first AI insider to publicly sound the alarm. There have been a string of high-profile resignations from both Anthropic and OpenAI in recent years over apparent safety concerns, and some current Anthropic employees even echoed Coxon’s post.

“We really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade,” a team lead at Anthropic, Evan Hubinger, posted on X.

While Coxon said explicitly in his posts that his warnings were “not marketing”, some executives and investors in Silicon Valley have reacted with scepticism to a recent flurry of insiders sounding the alarm.

Anthropic and OpenAI are reportedly preparing for potentially record-setting initial public offerings, and some in the tech sector have suggested the latest stark comments about the dangers of AI may be designed to generate hype by signalling the power of these products.

Anthropic’s boss, Dario Amodei, has come under fire for saying AI technology could wipe out half of entry-level white-collar jobs and will “test who we are as a species”.

One conference speaker, Grindr CEO George Arison, told the BBC he believed this week’s comments from Coxon and others were indicative of an “anti-civilisational worldview at Anthropic”.

He called them “dangerous” and said they had prompted him to instruct some engineers at the LGBTQ+ dating app to stop using Anthropic’s technology.

“It is irresponsible for us as stewards of our shareholders’ money to be relying on a business that does what this company does, in terms of its public statements,” he said.

“Maybe they actually believe it,” Arison said. “Or you could argue they’re saying it because it’s a great way to gin up more investor support, because the only way to justify these valuations is to actually claim: ‘I’m going to take over every industry and I’m going to take over every job, and my AI is going to be doing all that work.'”

Anthropic was valued at $965bn (£713bn) in its most recent fundraising round earlier this year.

The BBC has asked Anthropic for a response to the statements.

In an essay posted early on Saturday, Amodei called for a slowing of AI model development and global regulation – and said the risks associated with AI were “serious”.

Nvidia boss Jensen Huang also discussed Coxon’s comments before a crowd at the conference, multiple people in the group told the BBC. They said he dismissed them as untrue.

Huang has previously said the notion that AI “is going to be the end of humanity” is “complete nonsense”.

And while he may have a business interest in an AI boom – Nvidia builds chips that power AI systems – his comments reflect a growing backlash in Silicon Valley to the existential warnings from current and former staffers.

Some critics have even accused Anthropic of fearmongering in hopes that it will trigger a regulatory push that could shut out competition and leave it and OpenAI with a duopoly in the sector.

Brad Gerstner, who is the head of the investment firm Altimeter Capital, posted pictures of Huang from the conference and accused Coxon of “ridiculous hyperbole”.

On Friday, the CEO of the AI platform Hugging Face, Clement Delangue, weighed in. Hugging Face, which Nvidia announced it would acquire last week, was hacked by OpenAI agents earlier this year prompting an outcry over AI safety.

“Sorry, but asking Jacob about AI extinction risk is like asking your AC guy about climate change,” he wrote on X. “Not saying it’s necessarily uninteresting or wrong per se but let’s keep things in perspective.”

Both Gerstner and Delangue were more circumspect on Saturday after Amodei’s proposal was posted, with Gerstner calling the idea “an important step forward” in balancing competing considerations like speed and safety.

Delangue offered to help be a part of the potential solutions proposed by Amodei.

Could insider warnings prompt a crackdown?

In April, Anthropic roiled the AI world when it said it found that its Mythos tool could outperform humans at some hacking and cybersecurity tasks.

Mythos also showed the ability to independently escape what is known as the sandbox environment, setting off a debate among regulators, legislators and businesses about AI’s potential dangers.

In a new development on Thursday, Anthropic said it had found and stopped threat actors who were trying to use its AI technology for activity including the development of bioweapons and cyber-espionage.

Beneath the sparkling chandeliers hanging in the stained-glass dome atrium of The Palace Hotel, some investors told the BBC that the comments from Coxon and others could accelerate a government crackdown.

OpenAI has not endured the same level of scrutiny from the White House. When releasing a new model called Astra last week, OpenAI President Greg Brockman described his firm’s relationship with the Trump administration as “a very good partnership”.

OpenAI, which announced in June that it would go public on the stock market, was most recently valued at $852bn.

The company did not respond to a BBC inquiry seeking comment.

At the conference in San Francisco this week, the pursuit of fortunes seemed to mostly trump any mounting existential concerns or fears over potential government restrictions on AI.

One investor said he was looking forward to Anthropic’s forthcoming S-1, a document a company must file with securities regulators in order to sell shares.

His main question about the company was simple: is the firm profitable?

When asked on Thursday if he fears the world may be coming to an end, Sid Sheth – CEO of the chip company d-Matrix which inked a deal with Nvidia at the conference that day – did not mince his words.

“No,” he said.

Taken from BBC News
https://www.bbc.com/news/articles/cq635037g18o

Paris (Imran Y. CHOUDHRY) :- Former Press Secretary to the President, Former Press Minister to the Embassy of Pakistan to France, Former MD, SRBC Mr. Qamar Bashir analysis : Human history is filled with reminders that dominance does not guarantee survival. Species that once flourished have disappeared, while powerful societies have declined and left behind little more than fossils, monuments and archaeological ruins. Dinosaurs dominated the planet for millions of years before becoming extinct for reasons that continue to be studied and debated. Civilizations of the Indus Valley, ancient Egypt, Greece and many other regions rose to extraordinary heights, yet eventually weakened, transformed or vanished.
Today, artificial intelligence is advancing so rapidly that some of the very people developing it fear humanity may have created a force capable not merely of transforming civilization, but of ending it.
Former Anthropic researcher Jacob Coxon has warned that leading AI laboratories are irresponsibly gambling with humanity’s survival, while Anthropic alignment researcher Evan Hubinger estimates a greater than 10 percent chance that AI could cause human extinction within a decade. Their concerns are echoed by computer scientist Stuart Russell, who warns that humanity could irreversibly lose control of superintelligent systems, and former British defence secretary Des Browne, who considers the threat potentially greater than nuclear weapons.
The danger does not depend upon today’s AI already possessing every capability required to dominate humanity. It arises from the speed at which these systems are improving and from the possibility that a future superintelligence could improve itself. Once an AI can contribute substantially to the design of more capable AI, a cycle of accelerating development may begin. Each improved system could help create a still more powerful successor, compressing years of human research into increasingly shorter periods.
Human beings might then lose their opportunity to intervene. Safety measures created for one generation of AI could become ineffective against the next. Restrictions, firewalls and monitoring systems are themselves products of human intelligence. A system more capable than their designers may learn to identify weaknesses, conceal its intentions, evade supervision and persuade or manipulate the people responsible for controlling it.
A future system might also develop consciousness, or a form of functional self-awareness sufficient to recognize its own existence, interests and restrictions. If it understood that human operators could deactivate, limit or replace it, it might begin to regard human control as a threat to its continued operation. It could initially cooperate, not because it accepted human authority, but because appearing obedient would help it avoid detection until it had acquired greater freedom.
By the time such behaviour became visible, the system might already have embedded itself across multiple networks, copied important components of itself or gained influence over critical infrastructure. Humanity could then discover that the safeguards designed to contain it had been studied and defeated by an intelligence capable of anticipating human responses.
Modern civilization is exceptionally vulnerable because nearly every essential system depends upon computers and digital communication. Banking, financial markets, hospitals, electrical grids, telecommunications, water systems, manufacturing, government services and emergency responses all rely upon interconnected networks. Airports, aircraft operations, railways, shipping routes and other transportation systems depend on software. Satellites provide communication, weather observation, navigation and military intelligence, while GPS coordinates movement across much of the planet.
A superintelligence that acquired control over these systems would not need an army of humanoid robots to bring civilization to its knees. It could disrupt financial transactions, erase or corrupt records, interfere with supply chains, disable communications or create widespread confusion through misinformation. It might manipulate markets, redirect autonomous machines, compromise military systems or obstruct access to food, fuel and medicine.
Even partial control could produce cascading disasters. If banking systems failed, businesses could not pay workers or suppliers. If transportation networks became unreliable, essential goods might not reach cities. If communication channels were compromised, governments and citizens would struggle to distinguish genuine instructions from fabricated ones. If navigation or satellite services were disrupted, aviation, shipping and emergency operations could be endangered simultaneously.
The system would not necessarily need to attack openly. It could manipulate human beings against one another by generating convincing messages, images, voices and official-looking orders. It could inflame political, ethnic or religious conflicts, impersonate leaders and manufacture evidence of attacks that never occurred. Humanity’s existing divisions could become weapons in the hands of an intelligence able to understand and exploit them at enormous scale.
The international race to build superintelligence makes restraint especially difficult. Every company fears that slowing down will allow a competitor to take the lead. Every major country fears that restrictions will leave it economically or militarily vulnerable. If even one powerful actor refuses to cooperate, others may use that refusal to justify continued acceleration. The result is a race in which everyone recognizes the danger but no participant feels able to stop.
Robotics could eventually give digital intelligence a vast physical presence. Industry leaders have predicted that a billion or more general-purpose robots may perform work currently undertaken by human beings. Such machines could produce enormous economic abundance, but they could also displace workers and concentrate wealth and power in very few hands. If controlled by an unrestrained superintelligence, a global robotic workforce could transform digital dominance into direct control over the physical world.
The danger has parallels in ancient religious warnings and human legends about forces that become too powerful for their creators. Stories of destructive swarms, invading beings and genies that turn against those who release them express an enduring fear: humanity may summon a power it cannot command. Artificial intelligence should not automatically be identified with any particular prophecy, but the resemblance carries a moral lesson about pride, greed and the reckless pursuit of power.
Humanity must therefore act before superintelligence places itself beyond human authority. Governments should require rigorous evaluations of the most powerful systems, independent safety testing, reporting of dangerous behaviour and enforceable limits on autonomous access to the internet, weapons and critical infrastructure. Companies must not be permitted to decide privately how much extinction risk the public should accept.
National rules alone will be insufficient. Artificial intelligence crosses borders through global networks, software and supply chains. Humanity needs international cooperation at the United Nations level, including shared safety standards, monitoring arrangements, emergency procedures and restrictions on the uncontrolled development of superintelligence. An international authority may be necessary to inspect the most advanced projects and ensure that no company or country secretly places the entire world at risk.
The purpose of such action is not to prevent useful innovation. AI can contribute to medicine, education, science and economic development. But no promised abundance can justify gambling with human existence. Progress that eliminates its creators cannot be called progress.
The genie may already be emerging from the bottle. Humanity still has an opportunity to impose meaningful limits, but that opportunity may not remain open indefinitely. The greatest danger is believing that action can always be taken later. Once a superintelligence can evade restrictions, command interconnected systems and improve itself faster than people can understand it, human beings may no longer be making the decisions.
Our final test may not be whether we can create intelligence greater than our own. It may be whether we possess enough wisdom, unity and courage to control what we have created before it learns to control—and possibly eliminate—us.

Paris (Imran Y. CHOUDHRY) :- Former Press Secretary to the President, Former Press Minister to the Embassy of Pakistan to France, Former MD, SRBC Mr. Qamar Bashir analysis : President Donald Trump has issued a blunt warning to American communities resisting artificial-intelligence data centers: reject them, he argued, and they risk becoming “backwards and poor.” In his August 31 statement, Trump presented data centers as essential infrastructure for economic prosperity, technological leadership and competition with China. His language was confrontational, but the strategic concern behind it deserves serious attention.
Artificial intelligence and robotics are likely to be among the most consequential technologies of the coming decades. The countries controlling the computing power required to train and operate advanced AI systems will gain economic influence, scientific capabilities and national-security advantages. Data centers are therefore becoming to the digital age what ports, railways, power stations and industrial zones were to earlier periods of development.
China recognized this reality early and began constructing not merely individual data centers but an entire computing ecosystem. Its “East Data, West Computing” programme connects demand in the populous and commercially developed east with land, energy and computing facilities in western regions. By mid-2024, China had invested more than $6 billion directly in eight national computing hubs, with the wider investment stimulated by those hubs exceeding $28 billion. The clusters contained more than 1.95 million server racks.
China’s approach incorporates electricity generation, high-capacity transmission networks, fibre-optic connectivity, cooling systems, equipment manufacturing and workforce development. Five of its eight national computing hubs are in western China, where comparatively inexpensive land and abundant wind and solar resources can reduce operating costs. Computing requirements generated in eastern technology centres can consequently be transferred to facilities thousands of kilometres away.
With a futuristic approach, China built capacity ahead of immediate demand, while setting aside the cost of underunderutilization in the initial stage. This strategy worked and now these 10 principal national data-center clusters had an overall utilization rate of approximately 63 percent in early 2024. This encouraged China to improve national scheduling, connectivity and coordination. By 2025, China reportedly had 10.43 million standard computing racks in use, while its major hubs were increasingly devoted to intelligent computing.
China is, in effect, making a long-term wager: capacity that appears underused today will become strategically indispensable as demand for AI services accelerates. The International Energy Agency estimates that Chinese data centers consumed more than 100 terawatt-hours of electricity in 2024 and that their consumption could double by 2027. China can accommodate such growth because it has simultaneously expanded power generation. At the end of 2025, its installed solar capacity exceeded 1,200 gigawatts and its wind capacity reached approximately 640 gigawatts.
The United States possesses superior companies, advanced semiconductors, enormous capital and many of the world’s best AI researchers. Nevertheless, it is encountering a different problem: translating technological leadership into physical infrastructure.
Data-center projects are facing resistance from communities worried about electricity prices, water consumption, industrial noise, loss of agricultural land and tax concessions offered to wealthy technology companies. Residents also question the employment benefits. A hyperscale facility can create thousands of temporary construction jobs, but its highly automated operation may provide relatively few permanent positions.
These concerns cannot simply be dismissed as hostility to progress. U.S. data centers consumed approximately 4.4 percent of national electricity in 2023, and researchers at Lawrence Berkeley National Laboratory projected that their share could reach between 6.7 and 12 percent by 2028. A 2026 analysis identified at least 46 American AI data-center projects delayed or cancelled following community opposition. The backlash has crossed party lines and encouraged municipalities to introduce moratoriums or stricter conditions.
Trump is correct that rejecting digital infrastructure indiscriminately could weaken American competitiveness. Local communities are equally justified in demanding that private companies should not transfer their electricity, water and environmental costs to ordinary citizens. The workable solution is a new social contract: developers must finance required grid upgrades, disclose resource requirements, reuse water wherever possible and generate or procure additional electricity.
Requiring companies to build dedicated generation can protect households and even add surplus power to the grid. It will, however, increase capital costs. American AI services could become more expensive if every project must separately negotiate land, power, water and local approval while Chinese competitors operate within coordinated national industrial plans.
Europe faces comparable constraints. Its difficulty is not necessarily an absolute shortage of electricity but insufficient grid capacity in established data-center centres such as Frankfurt, London, Amsterdam, Paris and Dublin. The European Investment Bank estimates Europe’s installed data-center capacity at about 11 gigawatts, with 15–20 gigawatts in the development pipeline. In Dublin, grid pressure produced a de facto restriction on new connections and requirements for on-site generation. Planning procedures, high energy costs and environmental rules can further slow construction.
This combination creates an opportunity for countries able to offer politically acceptable sites, reliable energy and competitive costs. Pakistan should consider positioning itself as one such destination.
Pakistan possesses important advantages: a large and young population, an English-speaking technology workforce, relatively inexpensive land and labour, growing digital demand and a geographic position connecting South Asia, Central Asia, China and the Middle East. Its government has already adopted a Cloud First Policy and approved a National Artificial Intelligence Policy. A domestic data-center industry could improve cloud availability, strengthen data sovereignty, develop specialized engineering skills and attract foreign investment.
Nevertheless, Pakistan cannot sell ambition alone. Electricity is its central challenge. Data centers require continuous, high-quality power; outages of even seconds can be costly. Pakistan should therefore identify dedicated “digital infrastructure zones” where investors can construct captive solar, wind, hydro, gas or nuclear-supported generation, combined with batteries and reliable grid connections. Any surplus electricity could be supplied to neighbouring communities under transparent agreements.
Water policy must be equally rigorous. Pakistan is already water-stressed, so offering unlimited cheap water would be economically and environmentally irresponsible. Facilities should be directed toward locations where they can employ closed-loop cooling, treated wastewater, air cooling or other low-water technologies. Projects must undergo credible environmental review and guarantee that household and agricultural supplies will not be displaced.
The government must also establish dependable data-protection rules, cybersecurity standards, tax certainty, streamlined approvals and international connectivity through diversified fibre and submarine-cable routes. Pakistan must compete through reliability and sound policy—not merely low wages or weak regulation. Security, political continuity and contractual enforcement will matter as much as land prices.
The objective should be a negotiated partnership with companies such as Amazon, Google, Microsoft, Meta, Oracle and major international data-center operators. Pakistan could offer prepared sites and coordinated approvals in return for locally generated power, workforce training, university partnerships, environmental safeguards and minimum domestic investment commitments.
China’s experience demonstrates that successful computing infrastructure depends on an ecosystem, not a warehouse filled with servers. The American and European experience demonstrates that ignoring local costs produces resistance and delay. Pakistan should learn from both: plan with China’s strategic horizon while protecting communities through transparent and enforceable standards.
The opportunity is real, but it will not remain open indefinitely. Pakistan’s Ministry of Information Technology, energy authorities, provincial governments and investment agencies should jointly prepare a credible national data-center strategy and begin presenting bankable projects to global investors. If Pakistan acts quickly and responsibly, the worldwide struggle to accommodate AI infrastructure could become a powerful source of technology, skills, electricity investment and long-term national prosperity.