• Super Micro Computer shares surged 40% this week, reaching their highest level since February
• SMCI announced a $20 billion partnership with Saudi Arabian data center company DataVolt
• Raymond James named Supermicro a “market leader in AI-optimized infrastructure”
• The stock broke out above a falling wedge pattern on high trading volume
• SMCI is recovering from accounting issues that nearly led to a Nasdaq delisting

Super Micro Computer shares have rocketed upward this week, gaining a massive 40% in just a few trading sessions. The server maker’s stock closed at $45 on Wednesday after jumping 16% for the second straight day.

This dramatic price action comes after months of volatility for SMCI. The stock has been on a wild ride following accounting troubles that almost resulted in delisting from the Nasdaq.

Despite these challenges, Supermicro shares have climbed nearly 50% since the start of the year. Even more impressively, they’ve rebounded 63% from their April low.

The most powerful catalyst driving this week’s surge is a newly announced $20 billion multi-year partnership with Saudi data center giant DataVolt. This deal was revealed late Tuesday.

The partnership aims to accelerate the deployment of GPU platforms and rack systems across hyperscale campuses in the region. It represents a major expansion of Supermicro’s global footprint.

Impressive Technical Breakout

From a technical analysis perspective, Supermicro shares staged a decisive breakout above the upper trendline of a falling wedge pattern on Tuesday. The stock then accelerated above its 200-day moving average in Wednesday’s session.

Wednesday’s jump occurred on the highest trading volume since late February. This indicates strong buying conviction from larger market participants.

The relative strength index confirms bullish price momentum. However, it’s approaching overbought levels, which suggests the possibility of short-term pullbacks.

Traders are watching the psychological $50 level as the first overhead resistance area. This coincides with last August’s swing low and the upper range of an October consolidation period.

If the stock can close above $50, the next target would be around $63. This area matches the August countertrend high and prominent February peak.

For potential retracements, the $35 level serves as initial support. This aligns with Monday’s breakout point and corresponds to price action stretching back to last October.

Lower support sits around $26, near the early-February swing low. This could provide opportunities for longer-term investors.

Analyst Enthusiasm Adds Fuel

Wall Street’s growing enthusiasm has added more fuel to the rally. Raymond James recently initiated coverage with an Outperform rating and a $41 price target.

The analyst firm called Supermicro a “market leader in AI-optimized infrastructure” with pricing power that compares favorably to peers.

Adding to the bullish case, Supermicro revealed it had begun shipping high-density servers powered by AMD’s latest EPYC 4005 chips. This enhances the company’s AI hardware story.

The Saudi deal was announced as President Donald Trump began a four-day trip to the Middle East. Supermicro wasn’t alone in securing partnerships, as Nvidia and AMD also announced deals with firms in the kingdom.

Short interest in SMCI remains high at 21%, suggesting this recent run-up could partly reflect a short squeeze. The stock is still down more than 50% from its 52-week high.

While the company maintains its accounting issues are resolved, auditor BDO issued an adverse opinion on internal controls as of June 2024.

Looking ahead, investors should watch whether this breakout has staying power or if profit-taking emerges at key resistance levels.

The $20 billion Saudi partnership represents a long-term opportunity that could reshape Supermicro’s business trajectory. It also validates the company’s positioning in the competitive AI infrastructure market.

The post Super Micro Computer (SMCI) Stock: Server Giant Soars 40% on Saudi Deal and AI Momentum appeared first on CoinCentral.

• Pundi AI has been accepted into the NVIDIA Inception program for innovative AI startups
• The blockchain-AI hybrid platform offers omnichain data layer, tagging framework, and AI data marketplace
• The partnership provides access to NVIDIA hardware discounts, cloud credits, and training resources
• Pundi AI aims to create 1 million AI jobs globally through this collaboration
• The company’s token surged 30% following the announcement, reaching $0.15957

Pundi AI, a data infrastructure platform combining artificial intelligence with blockchain technology, has joined NVIDIA’s Inception program for cutting-edge AI startups. The move marks a rare crossover between the crypto and mainstream AI sectors, with Pundi AI gaining access to advanced tools and resources to accelerate its development.

The partnership, announced on April 29, represents an unusual acceptance for a crypto-adjacent project. NVIDIA’s Inception program typically excludes crypto-focused ventures according to its official guidelines, which bar digital asset businesses, consulting firms, cloud service providers, and public companies involved in cryptocurrency.

Pundi AI’s successful inclusion suggests the company has positioned itself primarily as an AI innovation platform rather than a crypto project. This strategic focus likely helped it qualify despite NVIDIA’s restrictive policy toward blockchain ventures.

Platform Capabilities and Vision

The company offers several key services that bridge AI and blockchain technologies. These include an omnichain data layer, a specialized framework for data tagging and annotation, and a marketplace dedicated to AI data trading.

At its core, Pundi AI aims to construct an open ecosystem for artificial intelligence. This vision focuses on democratizing access to data, development tools, and fostering decentralized innovation across the AI landscape.

The integration into NVIDIA’s program brings practical benefits for Pundi AI’s development. The startup will receive preferential pricing on high-performance computing hardware, cloud computing credits, and access to educational resources through the NVIDIA Deep Learning Institute.

These resources are expected to help Pundi AI reach important milestones faster. The company plans to use these advantages to speed up product development, enhance deployment strategies, and expand its user base.

Networking and Collaboration Opportunities

Beyond technical resources, Pundi AI gains entry into NVIDIA’s global startup community. This network connects emerging tech companies with industry veterans, creating opportunities for knowledge sharing and potential collaboration on advanced AI research and applications.

Zac Cheah, Pundi AI’s CEO and founder, emphasized the strategic importance of the partnership.

“Being a part of this ecosystem will help Pundi AI stay at the forefront of AI innovation, paving the way for impactful solutions that leverage NVIDIA’s high-performance infrastructure to transform industries,” Cheah stated.

Perhaps most ambitiously, Cheah highlighted that the partnership “marks a key step towards creating 1 million AI jobs globally.” This goal aligns with Pundi AI’s broader mission of expanding opportunities in the AI sector through its open ecosystem approach.

The announcement had an immediate market impact. Following the news, Pundi AI’s native token experienced a 30% surge in value, reaching $0.15957. This price movement reflects market optimism about the project’s potential growth with NVIDIA’s backing.

During critical development phases, Pundi AI will receive ongoing support from the NVIDIA Inception program. This includes customized resources such as additional NVIDIA Deep Learning Institute credits, exclusive hardware and software pricing, and technical assistance to drive the company’s growth trajectory.

The post Pundi AI Joins NVIDIA’s Elite Startup Program: Aims to Create 1 Million AI Jobs appeared first on CoinCentral.

• Sam Altman revealed that users saying “please” and “thank you” to ChatGPT costs OpenAI tens of millions of dollars
• 67% of American users are polite to AI assistants according to a December 2024 survey
• Debate exists over ChatGPT’s electricity consumption per query (between 0.3 and 3 watt-hours)
• OpenAI expects to triple revenue to $12.7 billion this year despite increasing competition
• The company doesn’t anticipate being cash-flow positive until 2029

OpenAI CEO Sam Altman has revealed that users being polite to ChatGPT is costing the company tens of millions of dollars. In a response on X (formerly Twitter) on April 16, Altman commented that the money spent processing “please” and “thank you” messages was “tens of millions of dollars well spent — you never know.”

tens of millions of dollars well spent–you never know

— Sam Altman (@sama) April 16, 2025

This revelation has sparked discussions across the tech community about the hidden costs of AI interactions. Each additional word in a prompt requires processing power and contributes to OpenAI’s operational expenses.

The statement highlights the scale at which ChatGPT is being used globally. With millions of daily users, even small courtesies add up to large expenditures for the company.

Why Users Remain Polite to AI

The disclosure prompted many to question why people feel compelled to be polite to AI systems in the first place. According to a December 2024 survey by Future, 67% of American users are polite to AI assistants.

Of those polite users, 55% reported doing so because they believe it’s the right thing to do. The remaining 12% admitted they were polite out of concern that mistreating AI could have future consequences.

Some users expressed that they interact politely with AI systems out of fear that if AI becomes sentient, it might treat humans based on past interactions. This perspective reflects growing public awareness of AI development and potential future capabilities.

Engineer Carl Youngblood offered a different motivation, suggesting that courtesy toward AI serves human development: “Treating AIs with courtesy is a moral imperative for me. I do it out of self-interest. Callousness in our daily interactions causes our interpersonal skills to atrophy.”

Technical Costs and Efficiency Improvements

The conversation around AI courtesy costs has led to debates about the energy consumption of ChatGPT queries. A September 2023 research paper by Digiconomist founder Alex de Vries claimed that a single ChatGPT query requires approximately three watt-hours of electricity.

This figure has been challenged by Josh You, a data analyst from AI research institute Epoch AI. You argues that the actual consumption is closer to 0.3 watt-hours, citing improvements in model efficiency and hardware since 2023.

In response to Altman’s post, some users questioned why OpenAI doesn’t implement solutions to reduce costs associated with processing courtesy words. Such technical adjustments could potentially save millions while preserving the user experience.

Altman has previously stated that the cost of AI output has been falling tenfold every year as models become more efficient. This rapid decline in operational costs suggests that the expense of processing courtesy words may become less burdensome over time.

Despite the current costs, OpenAI expects to more than triple its revenue this year to $12.7 billion. This growth comes amid increasing competition from companies like China’s DeepSeek.

However, the company does not anticipate becoming cash-flow positive until 2029, when it projects revenue will exceed $125 billion. This long-term outlook suggests that while courtesy costs are worth noting, they represent just one aspect of OpenAI’s complex financial picture.

The post Your “Please & Thankyou” to ChatGPT Is Costing OpenAI Millions: And They’re OK With It appeared first on CoinCentral.

• OpenAI is reportedly building an X-like social platform with AI image generation features
• The project is still in early development but has a prototype with a social feed
• CEO Sam Altman has been seeking feedback from outsiders about the concept
• This move intensifies rivalries with Elon Musk’s X and Meta’s planned AI assistant platform
• The development comes during ongoing legal battles between OpenAI and Musk

OpenAI is developing its own social media platform that could directly compete with Elon Musk’s X, according to reports from The Verge and confirmed by sources to CNBC. The project, still in its early stages, centers around a social feed integrated with ChatGPT’s popular image generation capabilities.

The move represents a new chapter in the growing rivalry between OpenAI CEO Sam Altman and Elon Musk. This development comes as the two tech leaders are locked in an intensifying legal conflict.

Sources familiar with the matter indicate that while the concept remains in prototype form, Altman has begun soliciting feedback from outsiders. The timing coincides with the massive popularity of OpenAI’s newest image-generation tools, which reportedly led to an overloading of the company’s servers.

“Our GPUs are melting,” Altman posted on X in late March, referring to the graphics processing units powering the AI workloads. The company temporarily limited usage of the feature while working to improve efficiency.

Strategic Expansion into Social Media

The potential social network would place OpenAI in direct competition with both Musk’s X platform and Meta, which is reportedly planning to add a social feed to its upcoming AI assistant app.

This positioning isn’t accidental. When reports of Meta building a ChatGPT rival first emerged a few months ago, Altman responded on X: “ok fine maybe we’ll do a social app.”

A source at another major AI lab told The Verge: “The Grok integration with X has made everyone jealous, especially how people create viral tweets by getting it to say something stupid.”

Developing a social platform would give OpenAI access to its own real-time user data, a valuable resource that X and Meta already leverage to train their AI models. Musk recently merged X and xAI into a single company, allowing Grok to surface content directly from X in its results.

The social media project’s existence shows how OpenAI is thinking about expansion at a time when expectations for future growth are high. The company recently closed what amounts to the largest private tech funding round on record, raising $40 billion at a $300 billion valuation in March.

Legal Battles and Power Struggles

The social media development unfolds against the backdrop of escalating legal conflicts between OpenAI and Musk. Last week, OpenAI countersued Musk, accusing him of attempting a hostile takeover via a $97 billion offer and using “bad-faith tactics” to gain control of the company he co-founded.

OpenAI is seeking damages and a court order to block further interference from Musk. When Musk made his unsolicited offer to purchase OpenAI in February, Altman responded on X: “no thank you but we will buy twitter for $9.74 billion if you want.”

In March, Musk filed a lawsuit against OpenAI, claiming it abandoned its nonprofit roots. While a judge denied Musk’s request to block OpenAI’s transition to a capped-profit structure, the case has been fast-tracked for trial in fall 2025.

OpenAI claims that Musk was actually the first to advocate for a for-profit model, as long as he controlled it. The company released internal emails showing Musk proposing in 2017 to take control of the board and become CEO, writing, “I would unequivocally have initial control of the company.”

Since leaving OpenAI, Musk launched xAI, which recently merged with X in an all-stock deal valuing xAI at $80 billion. The company’s flagship AI model, Grok-3, debuted in February and is expected to scale further.

It remains unclear whether OpenAI plans to release the social network as a separate app or integrate it into ChatGPT, which became the most downloaded app globally last month. An OpenAI spokesperson did not respond to requests for comment.

The project is reportedly inspired by the viral popularity of OpenAI’s image generation tools, particularly the anime-style renderings of users’ uploaded photos that have spread across X and other platforms. Altman himself changed his X profile photo to an image generated by this new feature.

OpenAI’s latest image-generation tool, released in March, creates everything from diagrams and infographics to logos and business cards. The feature can also use existing images as starting points for art, such as custom pet paintings or edited professional headshots.

The post AI Wars Heat Up: OpenAI Plans X Competitor as Legal Battle with Elon Musk Intensifies appeared first on CoinCentral.

• Meta received approval from EU regulators to train AI models using public data from adult users
• Public posts, comments, and interactions with Meta AI will be used for training, but private messages and data from minors are excluded
• EU users will be notified and given an opt-out option through a form
• The company had paused AI training last year after privacy concerns raised by Irish regulators
• Meta says this approach follows similar practices by Google and OpenAI in the European market

Meta has announced that it will resume training its artificial intelligence models using public data shared by adult users across its platforms in the European Union. This comes after nearly a year of paused operations due to data protection concerns from Irish regulators.

The social media giant will begin collecting public posts, comments, and interactions with Meta AI from adult users on Facebook, Instagram, WhatsApp, and Messenger for AI training purposes. The company emphasized that private messages between friends and family members will remain off-limits, as will any data from accounts belonging to users under 18 years of age.

EU users can expect to receive notifications this week through both in-app alerts and emails. These notifications will explain what types of data Meta plans to use and why this collection matters for improving AI systems and overall user experience.

Opt-Out Options Available

For those concerned about their data being used, Meta has promised an easily accessible opt-out mechanism. Users will find a link to a form where they can object to their public data being collected for AI training purposes. The company has committed to honoring all objection forms it has already received, along with any newly submitted ones.

“This training will better support millions of people and businesses in Europe, by teaching our generative AI models to better understand and reflect their cultures, languages, and history,” Meta stated in its announcement.

The European Data Protection Board (EDPB) recently approved Meta’s approach, confirming that it meets legal obligations under the EU’s data protection laws. This regulatory green light enables Meta to move forward with its AI training plans in the region.

Meta launched its AI assistant in the European Union last month across its major platforms. The company points out that it’s following similar practices already established by competitors like Google and OpenAI, who have also used European user data to train and fine-tune their AI models.

Privacy Concerns and Regulatory History

The resumption of AI training activities follows a challenging regulatory period for Meta in Europe. Last July, the company delayed its AI training plans after privacy advocacy group None of Your Business filed complaints in 11 European countries. These complaints alleged that Meta’s privacy policy changes would have allowed the company to use years of personal posts, private images, and online tracking data to train its AI products.

The Irish Data Protection Commission (IDPC) subsequently requested that Meta pause its rollout until a proper review could be conducted. Meta says it has now engaged “constructively with the IDPC” and received permission from EU data protection regulators.

Other tech giants have faced similar scrutiny in the region. The Irish data regulator opened a cross-border investigation into Google Ireland Limited last September to determine whether the company followed EU data protection laws while developing its AI models. X (formerly Twitter) agreed to stop using personal data from EU users for training its AI chatbot Grok after facing regulatory pressure.

The EU launched its AI Act in August 2024, establishing a legal framework for the technology that includes provisions for data quality, security, and privacy. This emerging regulatory landscape continues to shape how tech companies develop and train their AI models using user data in European markets.

Meta believes that diverse training data is essential for its AI models to understand the “incredible and diverse nuances and complexities that make up European communities,” including dialects, colloquialisms, and cultural context like humor and sarcasm.

The post Meta Receives EU Approval to Train AI Using Public Social Media Content appeared first on CoinCentral.

• Albert Saniger charged with fraud for falsely marketing Nate app as AI-powered
• The app actually used human workers in the Philippines to complete online purchases
• Saniger raised over $40 million from investors based on false AI claims
• The app’s actual automation rate was “effectively zero percent” according to authorities
• Nate ceased operations in January 2023 after media scrutiny

The founder and former CEO of e-commerce app Nate has been charged with fraud by U.S. authorities. Albert Saniger, 35, of Barcelona, Spain, allegedly misled investors by claiming his shopping app used artificial intelligence when it was actually powered by human workers in the Philippines.

The Justice Department and Securities and Exchange Commission (SEC) filed charges on April 9, 2025. Saniger faces one count of securities fraud and one count of wire fraud, each carrying a maximum sentence of 20 years in prison.

Founded around 2018, Nate was marketed as an AI-powered universal shopping cart that allowed users to complete online purchases with a single tap. The app promised to handle all checkout steps automatically, including entering shipping details and sizing information.

Tech CEO charged in artificial intelligence investment fraud schemehttps://t.co/UyoYN6lrtN

— US Attorney SDNY (@SDNYnews) April 9, 2025

Human Workers, Not AI

According to court documents, the reality was very different. Instead of using AI, Saniger employed “hundreds of contractors, or ‘purchasing assistants,’ in a call center located in the Philippines to manually complete purchases occurring over the nate app.”

While Saniger had acquired some AI technology from a third party and hired data scientists to develop it, authorities claim the app never achieved the ability to consistently complete e-commerce purchases. The app’s actual automation rate was “effectively zero percent,” according to the indictment.

Saniger allegedly went to great lengths to hide this fact. He told employees to keep the automation rate secret, restricted access to the company’s “automation rate dashboard,” and provided false explanations for his secrecy, claiming the automation data was a “trade secret.”

Millions Raised on False Claims

Under the guise of investing in an AI-powered app, Saniger solicited more than $40 million from venture capital firms. Some reports indicate the total funding may have reached $50 million, with investors including Coatue, Forerunner Ventures, and Renegade Partners.

Acting U.S. Attorney for New York Matthew Podolsky alleged Saniger duped investors by “exploiting the promise and allure of AI technology to build a false narrative about innovation that never existed.”

During the busy holiday season in 2021, Saniger reportedly directed Nate’s engineering team to develop bots to automate some transactions, working alongside the human workers. This decision came despite his previous statements to investors that the company did not use “dumb bots.”

Company Collapse

Nate’s practices came under scrutiny following a 2022 investigation by The Information that revealed the company’s heavy usage of human contractors. According to the DOJ, Nate ran out of money and was forced to sell its assets in January 2023, leaving investors with “near total” losses.

Saniger terminated all of Nate’s employees after media reports began casting doubt on the app’s capabilities. His LinkedIn profile indicates he was no longer CEO as of 2023. He is currently listed as a managing partner at New York VC Buttercore Partners.

The SEC’s civil action seeks to ban Saniger from holding office in any similar company and return investor funds. FBI Assistant Director in Charge Christopher G. Raia said, “Saniger allegedly abused the integrity associated with his former position as the CEO to perpetuate a scheme filled with smoke and mirrors.”

The case is being handled by the Office’s Securities and Commodities Fraud Task Force and the Complex Frauds and Cybercrime Unit.

Nate is not the only startup accused of exaggerating AI capabilities. Other companies, including an “AI” drive-through software startup and legal tech unicorn EvenUp, have also reportedly used humans to do much of the work claimed to be done by artificial intelligence.

The charges against Saniger highlight growing concerns about misrepresentation in the tech sector, particularly around artificial intelligence claims, as investors pour money into the booming AI space.

The post Former Nate CEO Charged with Fraud: ‘AI’ Shopping App Actually Powered by Humans appeared first on CoinCentral.

• OpenAI filed a countersuit against Elon Musk, accusing him of “bad-faith tactics” and a “self-serving narrative”
• Emails allegedly show Musk pushed to abandon OpenAI’s nonprofit model, take majority equity, and become CEO as early as 2015
• OpenAI seeks compensatory and punitive damages, plus an injunction to prevent further “harassment and disruption”
• OpenAI recently secured $40 billion in funding led by SoftBank, valuing the company at $300 billion
• Musk’s rival AI startup xAI merged with X last month in an all-stock deal valuing xAI at $80 billion

OpenAI has launched a countersuit against Elon Musk, claiming the billionaire attempted a hostile takeover of the company he helped found. The AI research giant is seeking damages and an injunction to block what it describes as Musk’s “campaign of harassment and disruption.”

In court filings dated April 10, 2025, OpenAI presented emails allegedly showing that Musk himself was the first to push for the company to abandon its nonprofit structure. The company claims Musk wanted to convert OpenAI into a for-profit entity as early as November 2015, just months after its founding.

“These antics are just history on repeat—Elon being all about Elon,”

OpenAI’s official newsroom posted on X, sharing internal emails to support their claims. Decrypt has not independently verified these emails, and representatives for Musk’s companies did not immediately respond to requests for comment.

Elon’s never been about the mission. He’s always had his own agenda. He tried to seize control of OpenAI and merge it with Tesla as a for-profit – his own emails prove it. When he didn’t get his way, he stormed off.

Elon is undoubtedly one of the greatest entrepreneurs of our…

— OpenAI Newsroom (@OpenAINewsroom) April 9, 2025

The Push for Profit

According to the emails released by OpenAI, Musk questioned the company’s structure in November 2015. In a message to CEO Sam Altman, Musk allegedly wrote that a “standard C corp with a parallel nonprofit” would “likely align incentives better.”

By summer 2017, as OpenAI’s computing needs grew with its Dota 2 experiments, discussions about funding intensified. On July 13, 2017, Musk reportedly agreed a for-profit model might be necessary for the company’s survival.

Days later, he warned that China’s AI ambitions were another reason to “change course.” This led to what OpenAI describes as Musk’s power play in September 2017.

The company claims Musk pushed for “initial control” over OpenAI’s board in exchange for capital. He also allegedly demanded to become CEO and proposed a structure where he would appoint four out of seven board seats.

“I would unequivocally have initial control of the company,” Musk wrote, according to the emails. OpenAI further alleges that Musk directed his team to incorporate a for-profit shell company called Open Artificial Intelligence Technologies, Inc., planning to move OpenAI’s intellectual property under it.

Co-Founders Push Back

OpenAI’s co-founders rejected Musk’s terms. In what the company describes as a candid message, they warned that Musk’s proposed structure risked creating an “AGI dictatorship.”

Musk’s alleged response was blunt: “Discussions are over. I will no longer fund OpenAI.”

In January 2018, Musk reportedly proposed spinning OpenAI into Tesla, claiming it was the only way to raise the billions needed. “OpenAI is on a path of certain failure relative to Google,” he wrote, according to the countersuit.

The OpenAI team declined again, unwilling to become a Tesla subsidiary. By February 2018, Musk resigned as co-chair and parted ways with the company.

The Legal Battle Intensifies

Years later, Musk returned with a lawsuit against OpenAI. In March 2025, he sought to block its transition to a capped-profit structure. A U.S. judge denied the injunction but agreed to an expedited trial, set for fall 2025.

OpenAI’s countersuit now seeks to block what it calls Musk’s “sham $97 billion takeover bid” and is asking for compensatory and punitive damages to be determined at trial.

“Through press attacks, malicious campaigns broadcast to Musk’s more than 200 million followers on the social media platform he controls, a pretextual demand for corporate records, harassing legal claims, and a sham bid for OpenAI’s assets, Musk has tried every tool available to harm OpenAI,” the company stated in its filing.

Musk’s attorney, Marc Toberoff, responded to the countersuit in an emailed statement: “Had OpenAI’s Board genuinely considered the bid as they were obligated to do they would have seen how serious it was. It’s telling that having to pay fair market value for OpenAI’s assets allegedly ‘interferes’ with their business plans.”

Competing AI Ventures

Musk has since founded rival AI startup xAI, which merged with X last month in an all-stock deal valuing xAI at $80 billion.

OpenAI, meanwhile, announced a $40 billion funding round led by SoftBank, pushing its valuation to $300 billion—nearly four times that of Musk’s xAI startup.

“We’re getting ready to build the best-equipped nonprofit the world has ever seen,” OpenAI said in its countersuit. “The idea that we abandoned the mission is false. Elon’s own emails make that clear.”

The company now seeks an injunction to force Musk to cease his public attacks. “Musk’s continued attacks on OpenAI, culminating most recently in the fake takeover bid designed to disrupt OpenAI’s future, must cease,” the filing states.

According to recent reports, OpenAI’s ChatGPT assistant now has 500 million weekly users, and the company is on pace for nearly $13 billion in 2025 revenue.

The post OpenAI Countersues Musk, Seeks Damages for “Campaign of Harassment” and $97B “Sham Bid” appeared first on CoinCentral.

• Former Meta executive Sarah Wynn-Williams plans to testify before Congress that Meta threatened US interests while aiding China
• Wynn-Williams alleges Meta briefed the Chinese Communist Party on emerging technologies, including AI, starting in 2015
• She claims there’s a “straight line” from these briefings to China’s current use of Meta’s AI tools for military capabilities
• Meta denies these allegations, calling her testimony “divorced from reality and riddled with false claims”
• Despite being banned in China, Meta reportedly earns revenue from Chinese advertisers, making it Meta’s second-biggest market

Sarah Wynn-Williams, who served as director of global public policy at Meta (formerly Facebook) from 2011 to 2017, is set to testify before the Senate Judiciary Subcommittee on Crime and Counterterrorism about her experiences at the tech giant.

Her testimony will focus on accusations that Meta executives undermined US national security by helping China advance in what she calls the “artificial intelligence arms race.”

In prepared remarks obtained by news outlets, Wynn-Williams claims that Meta began briefing the Chinese Communist Party as early as 2015. These briefings allegedly focused on emerging technologies, including artificial intelligence.

“The explicit goal being to help China outcompete American companies,” Wynn-Williams states in her draft testimony. She suggests there is a “straight line” connecting these briefings to recent reports that China is developing AI models for military use based on Meta’s technology.

Meta strongly disputes these allegations. Company spokesperson Andy Stone stated that her testimony is “divorced from reality and riddled with false claims.”

Secret China Operations

Wynn-Williams alleges that Meta established a secret relationship with China under a program called “Project Aldrin.” This initiative was reportedly restricted to need-to-know staff members.

According to her draft testimony, the company constructed a “physical pipeline” connecting the United States and China. She claims Meta executives ignored warnings that this could provide backdoor access to the Chinese Communist Party.

“Meta executives ignored warnings that this would provide backdoor access to the Chinese Communist Party, allowing them to intercept the personal data and private messages of American citizens,”

she states. She adds that Congressional intervention prevented China from gaining access to US user data through this pipeline.

Wynn-Williams also claims that Meta provided the Chinese government with access to user data, including that of Americans.

“I watched as executives decided to provide the Chinese Communist Party with access to Meta user data — including that of Americans,” her testimony reads.

Business Despite Ban

Facebook itself has never officially launched in China. The social media platform remains banned in the country over censorship issues.

However, Wynn-Williams plans to highlight that Meta continues to do business with China. She points to corporate filings that reference revenue Meta receives from Chinese advertisers.

Her draft testimony claims China is now Meta’s second biggest market based on SEC filings from last year. She also alleges that Meta’s AI model, Llama, has “contributed significantly to Chinese advances in AI technologies.”

Meta’s spokesperson counters these claims, stating:

“While Mark Zuckerberg himself was public about our interest in offering our services in China and details were widely reported beginning over a decade ago, the fact is this: we do not operate our services in China today.”

Legal Battles and Whistleblowing

Wynn-Williams recently published a memoir titled “Careless People” about her experiences at Facebook. The book quickly rose to bestseller status despite legal challenges from Meta.

The company successfully obtained an emergency arbitration decision that temporarily stopped Wynn-Williams and her publisher from promoting the memoir or making disparaging statements against her former employer. Meta is reportedly seeking hundreds of millions of dollars in their case against her.

“This gag order was sought by a company whose CEO claims to be a champion of free speech,” Wynn-Williams states in her draft testimony. “The American people deserve to know the truth.”

In addition to writing her memoir, Wynn-Williams filed a whistleblower complaint with the Securities and Exchange Commission in March. She has also filed complaints with the Department of Justice.

The testimony has drawn interest from lawmakers, including Senator Josh Hawley, a Republican from Missouri who leads the Senate Judiciary subcommittee. Hawley joined a bipartisan group of lawmakers in early April to investigate Wynn-Williams’ claims.

Meta maintains that Wynn-Williams was fired eight years ago for poor performance, while representatives for Wynn-Williams have declined to comment on this claim.

The hearing is scheduled for Wednesday, where Wynn-Williams will face questions from committee members about her allegations against one of the world’s largest technology companies.

The post Meta Whistleblower Claims Company Aided China’s AI Advancements appeared first on CoinCentral.

• CoreWeave raised $1.5 billion in its IPO, pricing shares at $40 each after initially targeting $47-$55
• The stock opened at $39, below the offer price, before closing flat at $40, giving the company a $23 billion valuation
• Nvidia backed the IPO with a $250 million order and is both a supplier and investor
• Microsoft accounts for 62% of CoreWeave’s revenue, with analysts citing customer concentration as a risk factor
• The debut occurred amid broader tech market declines, with the Nasdaq down nearly 3% on the day

CoreWeave marked its first day as a public company on Friday with a muted debut on the Nasdaq. The AI infrastructure provider, trading under the ticker CRWV, closed at its $40 IPO price after opening slightly lower at $39.

The company raised approximately $1.5 billion in the offering, making it the largest tech IPO in the United States since automation software maker UiPath went public in 2021.

CoreWeave had initially targeted a price range of $47 to $55 per share, which would have valued the company higher and raised about $2.5 billion. However, market conditions and investor concerns led the company to downsize its offering.

“There’s a lot of headwinds in the macro,” CoreWeave CEO Michael Intrator acknowledged on CNBC’s “Squawk Box” on Friday. “And we definitely had to scale or rightsize the transaction for where the buying interest was.”

The IPO gave the company a valuation of $23 billion on a fully diluted basis. This represents a significant milestone for the AI sector, as CoreWeave is considered the first pure-play AI infrastructure company to go public during the generative AI boom.

Nvidia, which supplies the crucial graphics processing units (GPUs) that power CoreWeave’s services, placed a $250 million order in the offering. The chip giant is both a key supplier and investor in CoreWeave.

The company’s business model involves renting out access to hundreds of thousands of Nvidia GPUs to large tech and AI companies. These chips have become a highly sought-after resource in the race to develop artificial intelligence applications.

CoreWeave reported impressive revenue growth in 2023, with sales soaring 737% to reach $1.92 billion. However, the company also posted a net loss of $863 million during the same period.

Microsoft

Microsoft represents CoreWeave’s largest customer, accounting for 62% of its revenue last year. The heavy concentration on a single client has raised concerns among some investors about the company’s long-term prospects.

“The OpenAI contract… materially decreases the single client concentration, and we’ll continue to do that over the next several years,” Intrator told Reuters when asked about these concerns. CoreWeave recently signed a five-year contract worth $11.9 billion with OpenAI.

The company’s debut coincided with a challenging day for tech stocks. The tech-heavy Nasdaq fell nearly 3% on Friday, heading toward its worst quarterly performance since mid-2022.

Some analysts have suggested that CoreWeave’s performance could serve as a bellwether for investor sentiment toward the AI sector. “CoreWeave stands to be a bellwether for the AI industry as a whole — a must-watch stock as questions about return on investment grow ever louder,” Bloomberg Opinion columnist Dave Lee wrote.

The company began in 2017 as an Ethereum-focused crypto mining operation before pivoting to AI infrastructure. It completely shut down its mining business in 2022 following Ethereum’s upgrade that reduced rewards for miners.

CoreWeave operates with substantial debt, having raised almost $13 billion in debt financing. As of last year, the company had around $8 billion in debt, with plans to use about $1 billion of the IPO proceeds to repay some of these obligations.

The company doesn’t own its data centers, instead leasing its 32 facilities and some equipment, resulting in operating lease liabilities of $2.6 billion. This capital-intensive business model reflects the high costs associated with building AI infrastructure.

CoreWeave faces competition from major cloud service providers including Microsoft, Amazon, Google, and Oracle, all of which offer their own AI infrastructure services.

The IPO was underwritten by a syndicate of 18 banks, led by Morgan Stanley, JPMorgan, and Goldman Sachs. Prior to the offering, CEO Michael Intrator controlled 38% of CoreWeave’s voting power, while Nvidia held 1%.

Investors and industry observers are now watching closely to see if CoreWeave’s public debut will encourage other AI-focused companies to pursue IPOs. Data analytics company Databricks and chatbot maker OpenAI have both recently raised funds at high valuations and could be candidates for future public offerings.

The post CoreWeave (CRWV) Stock: $1.5 Billion IPO Makes History as Largest AI Infrastructure Public Offering appeared first on CoinCentral.

• Crusoe Energy is selling its Bitcoin mining business to NYDIG to focus on AI infrastructure
• The deal includes 425 modular data centers across 7 US states and Argentina, totaling 270 megawatts
• About 135 Crusoe employees will join NYDIG with no job eliminations
• Crusoe pioneered technology that converts waste gas from oil fields into electricity for computing
• AI has become the majority of Crusoe’s revenue as the company sees greater growth opportunities in this sector

Crusoe Energy announced on March 25 that it plans to sell its Bitcoin mining operation to New York Digital Investment Group (NYDIG). The deal includes Crusoe’s Digital Flare Mitigation business which has been operating since 2018.

The transaction will transfer 425 modular data centers spread across seven US states and Argentina to NYDIG. These facilities represent approximately 270 megawatts of power generation technology.

Around 135 Crusoe employees will join NYDIG as part of the agreement. No roles will be cut as a result of the transaction, with staff continuing to operate the business under NYDIG’s ownership.

Crusoe Energy was founded in 2018 with a mission to solve an environmental problem. The company pioneered a technology called Digital Flare Mitigation (DFM).

This innovative solution captures natural gas from oil fields that would otherwise be burned off into the air. It then converts this “stranded energy” into electricity to power data centers placed directly at oil field sites.

The company initially used this energy to mine Bitcoin. Reports suggest Crusoe’s operation accounts for about 1% of the world’s Bitcoin mining activity.

Over time, Crusoe expanded its business model. The company began using its energy to power AI workloads on graphics processing unit (GPU) clusters.

From Bitcoin to AI

“The AI business — it’s become the majority of our revenue,”

said Cully Cavness, co-founder and chief operating officer of Crusoe. The company now sees AI infrastructure as its primary growth opportunity.

Crusoe is currently developing what could become one of the most powerful GPU clusters in the world. The company is building a hyperscale data center campus in Abilene, Texas.

This facility will start with 206 megawatts of capacity. It is expected to grow to 1.2 gigawatts by mid-2026, which could set a speed record for new data center development.

Crusoe is also expanding its cloud platform. The company is offering on-demand access to high-performance GPUs and is already running AI workloads in Iceland using only geothermal and hydropower.

Environmental Impact

Since launching seven years ago, Crusoe’s Digital Flare Mitigation business has made a positive environmental impact. The company reports it has prevented 2.7 million metric tons of greenhouse gas emissions.

Crusoe has stopped nearly 22 billion cubic feet of natural gas from being flared. This is equivalent to taking about 630,000 cars off the road for a year.

The company’s approach to energy and computing has influenced the broader industry. Crusoe proved that data centers could be effectively located in remote areas near energy sources.

“Crusoe’s innovative approach to energy utilized for mining is uniquely complementary to NYDIG’s bitcoin custody, institutional trading and mining businesses,” said Chase Lochmiller, CEO and co-founder of Crusoe.

For NYDIG, the acquisition helps expand its role in supporting Bitcoin’s proof-of-work security. The company is a leading bitcoin and power firm valued at approximately $7 billion.

“It is critically important to keep the Bitcoin network secure, and at the lowest possible cost,”

said Ross Stevens, NYDIG founder and executive chairman. He claimed that “fiat currencies are collapsing against Bitcoin around the world.”

NYDIG and Crusoe have worked together for years on flare gas projects. They have experience hosting each other’s equipment, which helped build trust between the companies.

Tejas Shah, CEO of NYDIG, expressed excitement about the deal. “Crusoe has built an extraordinary bitcoin mining business by demonstrating remarkable innovation—bringing together the industry’s top talent to solve complex challenges and unlock untapped energy sources.”

After the transaction closes, Crusoe will become a major equity holder in the combined entity. The company will rank second only to Stone Ridge, NYDIG’s parent company, in ownership stake.

The shift mirrors a trend in the tech industry. CoreWeave, another company that started in crypto before pivoting to AI, is scheduled to hold its stock market debut soon with a potential valuation over $25 billion.

Terms of the Crusoe-NYDIG deal were not disclosed. The transaction remains subject to regulatory approvals and customary consents before final closing.

The post Crusoe Energy Sells Bitcoin Mining Operations to NYDIG, Pivots to AI Infrastructure appeared first on CoinCentral.