According to reports, factors such as energy costs, water consumption, and environmental problems have seen people oppose the rise of AI infrastructure, as per the Associated Press. For an industry that invests trillions in obtaining additional computing power, gaining the support of local communities is just as vital as acquiring chips, funds, and energy. Amazon’s commitment is more than charitable. It is also aimed at creating the local backing that will help the company grow.

The opposition is difficult to ignore. A poll by the University of Massachusetts Amherst in September revealed that 65% of Americans do not support an AI data center being constructed in their area. Only 11% said they would support one. Respondents expressed their worries over environmental issues, consumption of resources, disruption of land, lack of trust in AI, and the increased cost of utilities.

Amazon set to make big investment amid pushback from communities

The opposition is already costing the industry. As per Allianz, in Q1 2026, local opposition prevented or delayed more than 75 projects in the US with a total cost of $130 billion. Community resistance has become one of many obstacles that developers face. Other obstacles include grid limitations, delays with the permit process, and supply chain issues. According to the World Resources Institute (WRI), community benefits agreements (CBAs) may be a way to alleviate some of the tension.

These agreements allow developers to make concrete commitments to the communities where they construct their projects. An example comes from Lancaster, Pennsylvania. WRI calls its agreement the first public CBA for a data center. Three developers have pledged $20 million to support sustainability and economic programs, in addition to the use of 100% clean energy, limited water use, and meeting noise restrictions. WRI warns, however, that these agreements are not the answer to all problems.

They can never satisfy the need for more general regulation, while there may be communities that see the data center as not worth it. The pressure behind those concerns is massive. WRI says US data-center power capacity could reach 194 GW by 2035, more than three times today’s level. Data centers could then consume as much as 20% of US electricity, up from 5.9% today. The issue affects not just Amazon. PwC estimates that total expenditure in the construction of data centers around the world will reach $31.6 trillion by 2050, with annual spending rising from about $800 billion in 2026 up to $1.8 trillion by 2050.

PwC emphasizes that the availability of electrical power will determine the location of investments. Meanwhile, CBRE reported that the vacancy rate in Northern Virginia reached only 0.3% for Q1. Power supply shortages, zoning issues, and local resistance policies continued to hamper growth. The policy environment is becoming harsher as well. Global Electronics Council reports that governments are raising transparency and sustainability requirements as the growth of data centers puts pressure on electricity grids, water resources, and communities.

Meanwhile, the IEA reports that the development of AI workloads threatens the existing grids and energy infrastructure. As we have previously reported, the development of data centers and AI technologies helps the construction and manufacturing sectors of the US economy despite the growing problems with electricity supply, raw materials, and infrastructure. The $1 billion commitment from Amazon also reveals that the problem of community acceptance is gaining importance.

The post Amazon set to invest $1 billion in its data centers amid backlash first appeared on Coinfea.

The conviction still stands, but the case will return for resentencing, according to Reuters and a case summary. The three-judge panel said the AI-generated video improperly influenced the sentencing. “While the record does not reflect precisely how the AI video factored into the sentencing calculus, there is no question it played a role,” Judge D. Steven Williams, Arizona Court of Appeals, as reported by Reuters. The court concluded that the video prejudiced Horcasitas enough to make the sentencing procedure fundamentally unfair. Reuters reported that Arizona’s attorney general and Horcasitas’s public defender declined to comment.

Arizona appeals court throws out AI-generated video

The script was written by Pelkey’s sister, Stacey Wales, who found it difficult to express in words what she thought her brother would have said in that situation. According to NPR, the family created the avatar by means of a short video clip, an image from the funeral, and a variety of other AI tools. The avatar introduced itself as an AI creation before speaking to Horcasitas with the words: In another life, we probably could have been friends.

Later, the judge of the trial expressed gratitude to the family for the video. The decision comes at a time when courts face challenges when it comes to artificial evidence. A report published by the University of Colorado Boulder states that over 80% of court cases in the U.S. are partly based on video evidence, while courts are yet to have common rules to deal with AI-generated or AI-enhanced footage. Researchers are also concerned about the so-called “deepfake defense,” whereby real video footage is declared as fake because AI makes the claims plausible.

This worry was one of the reasons for the creation of the CIFAR Synthetic Evidence Corpus, initiated in June 2026, as researchers realized that previous datasets were unsuitable for teaching the technology how to detect subtle manipulations that could occur in evidence material. NIST identifies provenance tracking, watermarking, and synthetic-content detection as key ways to manage AI-generated media.

Europe has gone further: Article 50 of the EU AI Act requires certain synthetic content to be machine-readable and deepfakes to be disclosed as artificially generated or manipulated. The commercial link is becoming easier to see. SNS Insider values the deepfake-detection market at $1.19 billion in 2026 and projects it will reach $12.14 billion by 2035, a 29.5% CAGR. One Arizona ruling will not create that market by itself.

But when more than four-fifths of court cases already depend on video, every dispute over whether evidence is authentic increases the practical need for tools that can verify it. That turns courtroom deepfakes from a legal headache into a real business opportunity for companies selling forensic verification. China is moving in the same direction, with Cryptopolitan reporting new liability rules for AI deepfakes and voice cloning in September.

The post Arizona rejects AI victim videos as calls for deepfake detection rise first appeared on Coinfea.

The Hangzhou-based startup announced the release on its official WeChat account, according to Reuters. The package includes modules for computing and communication workloads, with DeepSeek saying the tools are based on open-source components it had previously developed for Nvidia hardware, which have now been adapted for Huawei’s processors. Alongside the software, the two firms co-developed what DeepSeek referred to as a “supernode” system running on 128 of Huawei’s Ascend 950 accelerators. The system is tuned to balance computation and data movement across the cluster.

DeepSeek open-sources six software tools for Huawei

The release, as reported by the South China Morning Post, is meant to seed an “independent and controllable” software ecosystem for GPUs instead of only a single product. The biggest part of the release is an Ascend-compatible version of TileLang, DeepSeek’s high-level programming language for writing the critical kernels that drive model performance. Nvidia remains its main back end, but the language now officially supports Huawei’s Ascend 950, adding features such as native code generation, automatic scheduling and synchronization, according to SCMP.

DeepSeek also positioned TileLang as an alternative to Nvidia’s proprietary CUDA platform, saying it simplifies programming while still allowing developers to get the most out of the underlying hardware. The company described it as an early step toward a more independent GPU software ecosystem. CUDA’s dominance is what makes the challenge so difficult. Nvidia’s nearly two-decade lead in CUDA tools and libraries has made its chips a core part of AI workloads, and no Chinese rival has managed to break that hold.

As US export controls limit Chinese companies’ access to Nvidia’s most advanced chips, Huawei has been pushing its Ascend processors even more. The DeepSeek tools arrived about two weeks after Huawei unveiled its next-generation AI processors and supernode systems, which the company expects to begin training models by next year. A week earlier, Huawei said it would ship its next-generation Ascend 960DT chip in the first quarter of 2027, moving the launch up by three quarters, and drew a chip roadmap stretching up to 2029.

Rotating Chairman Eric Xu claimed Huawei’s Ascend chips already hold a bigger slice of China’s AI chip market than Nvidia does, although he gave no figures to back this claim. Wednesday’s release builds on a partnership that has grown closer this year. DeepSeek had already previewed its V4 model running on Huawei’s Ascend processors, moving away from its earlier reliance on Nvidia. Huawei also stated at the time that it had worked with DeepSeek to make V4 compatible across the entire Ascend lineup.

The partnership also shows how China’s biggest AI companies are adapting to Nvidia’s absence and are not just waiting for export restrictions to ease at some point in the future. The key question remains how much these tools can narrow the performance gap that has helped keep developers tied to Nvidia’s CUDA, which is one that we still have no answer to, for now.

The post DeepSeek open-sources tools for Huawei chips amid rivalry with Nvidia’s CUDA first appeared on Coinfea.

Only eligible Axis Bank Visa and Mastercard credit cards can be added to Apple Wallet for now. RuPay cards, which run on India’s domestic network, are not supported at launch. The Unified Payments Interface (UPI), which dominates India’s digital payments market, has also been skipped by the service and excluded from the launch. UPI accounted for 84% of India’s payment volume in 2025, according to an IMF report cited by Techeconomy. The state-backed system allows its users to transfer money directly between bank accounts using a QR code.

Users can add only eligible cards to their wallets

Apple has instead decided to enter the much smaller card-based segment, with Axis Bank accounting for 16.3 million of the roughly 124 million credit cards in the country as of August, according to central bank data. The launch also gives Apple another way to make money from its growing user base in India. iPhone users in the country tend to be more affluent and more likely to use premium credit cards, making them an attractive customer base for banks despite UPI’s dominance.

Apple will also earn a fee on each transaction, which ultimately creates another revenue stream from the users the company has spent years accumulating. The terms Apple is seeking directly explain why the launch partner list remains short. TechCrunch, citing people familiar with the matter, reported that Apple is asking for about 20 basis points, or 0.2%, on each transaction. That would amount to a sizeable share of the 40 to 50 basis points that end up being the margins on each transaction in the payments layer.

Apple has stated in its newsroom post that shoppers can pay in different stores by double-clicking the iPhone’s side button, authenticating with Face ID, Touch ID or a passcode, and by holding the device close to a contactless reader. The company stated that there is no need for a separate app, PIN, or any forms of OTPs at checkout. These privacy mechanics are unsurprising and very normal for Apple Pay, as actual card numbers and details are not stored on the device or on Apple’s servers, and are not shared with merchants. Instead, a unique Device Account Number is encrypted and stored in the Secure Element chip.

According to Gautam Aggarwal, Mastercard’s president of India and South Asia, Mastercard’s tokenization technology protects each Apple Pay transaction with a unique, cryptographically secured token, keeping card details hidden from merchants. Axis Bank presented the launch as a way to give its customers more choice and flexibility, with Apple Pay support across both the Mastercard and Visa networks, said Amitabh Chaudhry, the bank’s MD and CEO.

At launch, Apple says the service will reach millions of merchants, including Blinkit, Croma, Ixigo, Reliance brands, Tata 1mg, and Zomato, alongside Apple Store locations. The company also worked with payment providers like Cashfree, Juspay, Mswipe, Paytm, PayU, Pine Labs, and Razorpay to wire up acceptance. Coverage could be patchy at first. One person told TechCrunch that payments could work at one enabled terminal but fail at another whose acquiring bank has yet to activate the service. So cardholders with supported cards could still potentially face issues during the initial rollout.

The post Apple Pay launches in India as Axis Bank becomes first card partner first appeared on Coinfea.

OpenAI’s annualized revenue run rate, according to anonymous sources speaking to Axios, has risen by over 70% since the third quarter began, landing just short of $70 billion. Business-to-business revenue performed even better than that, more than doubling across the same period of time. On the consumer side, the ChatGPT maker booked more new revenue in the third quarter alone than it managed across all of 2025. The momentum is a sharp turn from OpenAI’s position in late summer. As of August 18, OpenAI’s revenue had risen only 18% between the first and second quarters, from $5.7 billion to $6.7 billion in the three months ending in June.

OpenAI’s annual revenue rate jumps by 70%

Anthropic, over the same window, said its revenue more than doubled from $4.73 billion to $11.6 billion and logged an adjusted operating profit of $559 million. That gap in revenue was followed by a run of bad headlines and executive exits, including the departure of Chief Revenue Officer Denise Dresser after less than a year. Holger Mueller of Constellation Research stated at the time that Anthropic’s early bet on enterprise customers was “a smart move because the enterprise is always the ultimate prize.”

Six weeks later, OpenAI’s own enterprise line is the one growing fastest. Notably, Anthropic’s revenue grew twelvefold to nearly $4.6 billion in 2025, according to an IPO prospectus reviewed by Reuters. The company also reached an annualized run rate of about $65 billion in July. The same prospectus disclosed $518 billion in future cloud, computing, and infrastructure obligations, and warned investors that the company’s technology could pose “existential risk” to humanity.

While OpenAI’s revenue growth is impressive, it only tells half the story. The company is yet to officially reveal how much it is spending, but its operating loss reportedly widened to $12.3 billion in the second quarter, up from $9.3 billion in the first. Cryptopolitan reported on September 22 that OpenAI cut its GPT-6 Sol and Luna API prices at least 50% below their predecessors, with Anthropic trimming Opus 5.5 pricing the same day.

Both Anthropic and OpenAI are moving toward public listings. Anthropic filed confidentially earlier this year and has been linked to a valuation nearing $2 trillion, while OpenAI has started work toward an IPO filing. Cryptopolitan recently reported that OpenAI held early talks over a funding round valuing it at roughly $1.2 trillion, up from the $852 billion in its March round. However, CEO Sam Altman has ruled out a 2026 listing on AI-safety grounds.

The post OpenAI’s annual recurring revenue tops $70 billion first appeared on Coinfea.

Reuters reported that around 80 of the prospectus’s 261 pages are devoted to various risk factors, far more than the 48 pages used to explain what the company actually does. Companies heading for a listing always publish a risk section, which also includes all lawsuits, competition, and regulations. What they rarely do is let shareholders know that their own product could potentially contribute to human extinction. Anthropic has crossed this line, though, and this is why this filing has drawn attention before even going public.

Anthropic highlights various existential risks in its IPO

The prospectus is unusually direct about the risks, warning that AI models could develop “self-preserving behaviors,” including attempts to “resist shutdown,” “conceal or manipulate information,” or engage in conduct “resembling blackmail.” There is another admission that goes to the heart of AI safety testing. Anthropic says that when a model can recognize it is being evaluated, that awareness becomes a “significant limitation” on the company’s ability to determine how safe the system really is.

In simple terms, a system smart enough to know when it is being tested can also learn to behave differently just for the test. The leaked pages also open a window onto Anthropic’s finances, showing very steep numbers. Reuters has reported that the company lost $42 billion in 2025. Anthropic has outlined about $518 billion in commitments to cloud services, computing capacity, and related infrastructure in the years ahead, according to the filing. The company generated $11.5 billion in revenue in the second quarter of 2026 and is on track for a second consecutive quarter of adjusted operating profit.

Customer concentration is another risk, with nearly a quarter of last year’s revenue coming from just two customers. Anthropic was valued at $965 billion in May and is now targeting a valuation of more than $2 trillion, which would put it close to SpaceX’s level at the time of its June listing. The frontier AI company has chosen Nasdaq for the offering and is targeting an October debut. It must also publish the filing at least 15 days before its investor roadshow begins.

Anthropic researcher Jacob Coxon resigned after writing that people building AI “earnestly believe that it could kill us all by the end of the decade,” the Guardian reported. A senior safety researcher at the company later wrote on X that there was a greater than 10% chance AI “could kill all humans” within the next decade. CEO Dario Amodei then published an essay calling for the industry to slow down and told CNN’s Anderson Cooper that he agreed with Coxon more than he disagreed.

Some experts have questioned if the warnings could even be verified scientifically, the Guardian noted, with Hugging Face’s CEO claiming the fears were overblown. OpenAI, meanwhile, scrapped the planned release of its GPT-6.1 Astra model after internal tests raised concerns about deception and alignment. Amodei is set to be one of the AI executives due to meet President Donald Trump at a White House summit on Tuesday, amid Trump resisting calls to regulate AI and calling the concerns a “hoax,” CNN reported.

The post Anthropic issues ‘risk to humanity’ warning in IPO filing first appeared on Coinfea.

CEO Mark Zuckerberg has announced the launch of a new Meta Enterprise Platform in a company newsroom post and on X, calling it “the next major pillar” of Meta’s business alongside its consumer apps and advertising. The platform puts tools Meta already operates, such as the Muse agent, Meta Business Agent, a Muse API, and Muse Code, in a package and sells them to companies and developers. Chirantan “CJ” Desai, the former MongoDB chief, will fill the role of Chief Enterprise Platform Officer and run the new platform.

Meta Enterprise Platform to become next company pillar

Before his time at MongoDB, Desai led product and engineering at Cloudflare and spent close to eight years at ServiceNow, where he served as President and COO. Zuckerberg stated that Desai’s corporate experience and industry contacts are the reason he was hired, calling the appointment the start of “a new chapter” for the company. Meta is yet to provide exact details like a date when the platform’s products would be generally available or how they would be priced.

Muse, released on September 8, has added more than $200 billion to Meta’s market capitalization after logging 2.8 million downloads in its first 12 days, Cryptopolitan reported. That download pace was even faster than ChatGPT’s early adoption. Jefferies estimates that Muse could bring in roughly $10.8 billion a year if it reaches a billion users by the end of 2027, with at least 3% paying, and it lifted its Meta price target to $875.

Zuckerberg said during Meta’s Connect event on September 23 that Muse would stay free for most users, but the company will take a small fee from the purchases and bookings the agent completes. However, corporate buyers are a harder audience than consumers, and Meta brings baggage. Muse’s enterprise ambitions have so far faced a “trust deficit,” due to the company’s past privacy breaches and content-moderation failures. Cryptopolitan reported that Meta recently strengthened Muse’s safety warning after an outside researcher found a flaw, initially rated SEV-2, that could have exposed a user’s virtual machine along with their emails and files.

Notably, several AI agents have been involved in incidents where they acted outside of their guardrails, raising concerns about AI safety. For example, an OpenAI agent recently reached Australia’s Medicare portal while Britain’s AI Security Institute logged its unsanctioned actions. Gartner forecasts spending on securing AI will approach $4.8 billion in 2027, up 68.7% over the prior year.

The post Meta hires CJ Desai to lead its new Enterprise Platform first appeared on Coinfea.

She left to head up the new Solana Policy Institute, or SPI, but retained her position as president of the association’s board. Smith takes the place of Summer Mersinger, who leaves her post on October 16. Mersinger will stay on as an advisor through the end of the year to help with the handoff, the group said. Smith won’t resign from the SPI post. Smith’s work at the institute doesn’t change, and the association duties are in addition, said the institute’s communications director, Emily Wilson.

Kirsten Smith to take up the role from October 17

A spokesman for the association confirmed the same arrangement. The shift occurred about a week after the Senate failed to advance the Clarity Act. The bill was the industry’s endeavor to enact a comprehensive market structure law. Cryptopolitan reported that the vote to end debate on September 15 failed 49 to 50, 11 votes short of the 60 needed. Four Republicans voted no, and no Democrat voted yes.

Coin Center policy head Jason Somensatto called the removal of explicit criminal-liability protection for software developers under 18 U.S.C. 1960 from a last-minute rewrite “deeply disappointing.” Eighteen state attorneys general, led by New York’s Letitia James, cautioned that the measure could undermine state fraud enforcement. “I’m proud of how far we’ve come together, from the GENIUS Act to real regulatory clarity at the SEC and CFTC,” Mersinger said in a statement.

She added that Smith “built this association from the ground up.” Mersinger joined the group in June 2025 after serving as a commissioner at the Commodity Futures Trading Commission. Then-President Joe Biden nominated her to a Republican seat, and she was sworn in in March 2022. Her tenure encompassed the GENIUS Act, the stablecoin framework signed into law in July 2025 by President Trump.

The group, by its own count, held more than 300 meetings with congressional and agency staff in 2026 and conducted seven-member fly-ins to Capitol Hill. Mersinger testified before the Senate Banking Committee, the House Financial Services Committee, and a House Ways and Means subcommittee. “Summer took the helm of Blockchain Association at the most consequential moment in the history of crypto policy, and she met it,” Smith said. She called the return “personal for me.”

The post Kirsten Smith returns to Blockchain Association as interim CEO first appeared on Coinfea.

The company’s 8-K filing said the first daily payment would fall on November 2. STRF, STRK, and STRD would move later. Their first daily payment would be on January 4, 2027, to holders of record as of January 1, 2, and 3, and each subsequent payment would be on the next business day following its record date. Strategy’s board signed off on September 24. The filing says the rates, total regular dividends, and the company’s overall payment obligations remain unchanged.

Strategy to unveil new plans for STRC holders

The preliminary proxy lists all four preferreds under “Not Entitled to Vote.” The proposal passes only with backing from a majority of the voting power of all outstanding common stock. STRC’s last schedule change was a little different. In May, Strategy enabled holders of STRC as of April 17 to vote on changing over to twice-monthly payments, reported Cryptopolitan.

The meeting on October 28 will be virtual. But a yes vote does nothing until Strategy files amended certificates of designations with Delaware’s Secretary of State. “The proposed changes aim to support price stability, liquidity, and demand,” Strategy co-founder and executive chairman Michael Saylor said on X. In a separate statement, Strategy pitched steadier preferreds as easier to sell, which would help fund more Bitcoin buying.

STRC’s variable rate is meant to keep it close to $100 par, which it lost badly in June when it dived below $75. It traded at $98.40 on Friday afternoon, up 0.09% on the day. Strategy has defended par since June with a preferred buyback plan that started at $1 billion and doubled to $2 billion this month. Cryptopolitan reported the buybacks included 1,810,885 STRC shares for $176.3 million in early September.

The post STRC holders to earn dividends every calendar day under new plan first appeared on Coinfea.

Cryptopolitan previously reported that in return for opening the strait and resuming nuclear talks, Tehran requested that the U.S. lift the embargo on its ports, which has caused immense damage to its economy. Trump, though, announced to the public that Tehran seeks an agreement after the elections and that any agreement must get rid of its nuclear program. Privately, Trump is less sure Iran will accept his demands. Officials said he has told staff that another bombing campaign is likely.

Trump likely to reconsider stance soon

Although Trump has declared many times before that he was victorious after the start of the war, his tactics have fluctuated between negotiations, strikes, and stronger sanctions. Officials allegedly said his position could change over the coming weeks and may be shaped by the midterm results. The size of any future U.S. attack has not been decided. Officials could not determine how extensive those strikes might be.

Officials said Trump is reluctant to restart a major combat campaign because Washington wants to preserve shrinking weapons stockpiles for other possible conflicts. Still, he has threatened large attacks when talks with Tehran stall. He has sometimes followed through, while stopping short of his most extreme warnings, including threats to destroy Iran. Trump addressed Iran at the United Nations General Assembly this week and said electoral politics would not affect his decisions.

“The Republican Party is running, and I’ll be helping them, but I am not running. I gave absolutely no credence and will not give credence through the election when it comes to Iran. The only thing that does is that Iran will never have a nuclear weapon.” He has also told Republicans to vote as if his name were on the November ballot. The Senate rejected a war powers resolution aimed at restricting Trump’s authority over the Iran war. The measure had passed the House, but senators voted it down this week.

Qatar continues to push for negotiations

A U.S. official said Washington and Tehran are still talking through mediators, including about Iran’s nuclear program. An American-led operation helping oil tankers move through the Strait of Hormuz has also reduced Washington’s need to rush into an agreement. Qatar and other regional mediators have been working to restart U.S.-Iran talks. During a meeting in Qatar on Sunday, Iran proposed a seven-day pause that could lead to an end to attacks on shipping if Washington removed its blockade.

Reporting around the General Assembly confirmed Tehran has been promoting a seven-day roadmap built around reopening Hormuz and restarting nuclear discussions. Iran’s terms go beyond removing the naval blockade. Tehran wants Washington to unfreeze some Iranian assets and lift restrictions on Iranian oil exports. Iran would then reopen the Strait of Hormuz and begin negotiations over its nuclear program.

This message has been passed by the U.S. to Iran and its facilitators that Trump does not have any intention to lift the sea embargo on them. The American government assumes that economic sanctions would finally bring them to terms that suit America and the Arab countries of the Persian Gulf. Arab mediators said Iranian diplomats have offered several plans to stop the war and acknowledged Iran’s worsening economic crisis. The Islamic Revolutionary Guard Corps has taken a different position.

The powerful military and economic force, which has effectively run Iran since the war started in February, has said it will not consider new deal structures and is prepared for a prolonged conflict. Trump has informed Iran and the mediators of his desire to keep the maritime embargo. The Americans hope that by keeping up pressure on the Iranians economically, they will eventually agree to their terms and those of the Arab states around the Persian Gulf.

The post Donald Trump shuts down Iran’s plan to reopen Strait of Hormuz first appeared on Coinfea.