Artificial intelligence will be the biggest bubble of all time, according to the CEO of open-source AI company Stability AI.

Speaking with UBS analysts on a call last week, Stability AI CEO Emad Mostaque said of artificial intelligence: “I think this will be the biggest bubble of all time.” He added that it is still at the very early stages and not ready for mass-scale adoption in industries like banking just yet.

“I call it the ‘dot AI’ bubble, and it hasn’t even started yet,” he said.

Stability AI is the company behind Stable Diffusion, one of the other more popular generative AI tools aside from OpenAI.

Stable Diffusion allows users to generate photo-realistic images by inputting text. It has more than a million users and has raised over $100 million from investors including Coatue and Lightspeed Venture Partners.

Mostaque, its co-founder and CEO, has been accused of making misleading claims about his background, achievements, and partnerships. He disputed the claims one by one in a detailed response on his personal blog.

Generative AI has captivated the imagination of many an academic, boardroom executive, and even school student, for its ability to produce humanlike language and visual content from scratch in response to user prompts by using vast amounts of data.

AI has long been around, with the technology now a common feature of online browsing, social media platforms, and home assistants. Beyond consumer applications, the technology is being used in medicine, transportation, robotics, science, education, finance, defense, and other industries.

However, a more novel form of AI which has come about recently is generative AI, which is used in tools such as the popular AI chatbot ChatGPT, from U.S. tech firm OpenAI, as well as Google Bard and Microsoft Bing Chat, and image generators like Dall-E, Stable Diffusion, and Midjourney.

Mostaque said that the total amount of investment needed in AI was likely to be $1 trillion “because it’s more important than 5G as infrastructure for knowledge,” and suggested banks like UBS would have to adopt the technology as it is a “massive market.”

But, he added, it is at the “early stages” of development right now.

“It’s not quite ready” to be deployed at scale within large industries like financial services, “but we can see the value,” Mostaque said.

Mostaque said that companies that do not use AI appropriately in their businesses will be “punished” by the stock market.

He cited the example of Google, which lost $100 billion in a single day after its Bard AI chatbot gave inaccurate information in a promotional video upon its release. Google is competing aggressively with Microsoft to win in the race to build superior AI tools.

“I think this is real. I think that there aren’t many investable opportunities here, and you’ll see people moving from the best chip manufacturers to companies that are using this to impact their bottom line and their top line appropriately. And you will see the market punishing those that don’t use this,” Mostaque said.

“This will be one of the biggest investment themes over the next few years,” he added.

WATCH: Inflection A.I.’s Reid Hoffman: A.I. is the new industrial revolution, the ‘cognitive revolution’

SoftBank Group chairman and CEO Masayoshi Son on Wednesday said that the Japanese investment firm plans to shift from “defense mode” to “offense mode” and wants to capitalize on the AI revolution.

“Now, the time has come to shift to offense mode,” Son said during a shareholders’ annual general meeting.

“In the past few years, we focused on being [on] ‘defense.’ Three years ago, we didn’t have a lot of cash on hand. But because we have been in defense mode, we have built our cash on hand to five trillion yen ($35.3 billion),” Son said.

“We are ready to shift to offense mode. I am excited about that,” said Son.

The tech conglomerate, which engages in venture capital investing through its Vision Fund, has had its fair share of ups and downs. It was in “defense mode” as it halted new investments and trimmed its stake in Alibaba. In May, the Vision Fund reported a record $32 billion loss.

“What I am interested in most, what I am working on most, is the AI revolution. I believe that mankind is going to be exceeded by computer or AI,” said Son.

“We would like to be [in] the leading position for the AI revolution,” said Son.

SoftBank shares rose 2.63% in Wednesday morning trade.

The Vision Fund has invested in Chinese tech firms and therefore was hit by Beijing’s crackdown on the country’s tech sector and the subsequent plunge in share prices. SoftBank’s portfolio companies include ByteDance, DiDi Grocery, Coupang and more.

SoftBank is gearing up for the IPO of Arm, the U.K.-based chip design firm it acquired in 2016. Arm filed for the listing in the U.S. SoftBank’s CFO Yoshimitsu Goto said the IPO process is “going smoothly.”

Artificial intelligence has seen explosive growth in recent months, fueled by chatbot ChatGPT’s virality. ChatGPT has amazed researchers and the general public with its ability to generate humanlike responses to users’ prompts.

Son said on Tuesday that he is “a heavy user of ChatGPT” and that ChatGPT is “amazing.”

“The fortunes of SoftBank [are] looking to turn,” said Amir Anvarzadeh, Japan equity market strategist at Asymmetric Advisors on CNBC’s “Street Signs Asia” Wednesday after SoftBank’s shareholders’ meeting.

“You can see why Nvidia wanted to buy Arm a few years back because obviously, they wanted all of the architecture to themselves. Now it kind of makes sense, looking back.” The U.S. chip maker had dropped out of the $40 billion deal to acquire Arm.

“$30 billion is what we thought could be what Arm is worth. I reckon now, even $60 billion, might not seem too insane, given the backdrop,” said Anvarzadeh.

LISBON, Portugal — Once high-flying tech unicorns are now having their wings clipped as the era of easy money comes to an end.

That was the message from the Web Summit tech conference in Lisbon, Portugal, earlier this month. Startup founders and investors took to the stage to warn fellow entrepreneurs that it was time to rein in costs and focus on fundamentals.

“What’s for sure is that the landscape of fundraising has changed,” Guillaume Pousaz, CEO of London-based payments software company Checkout.com, said in a panel moderated by CNBC. 

Last year, a small team could share a PDF deck with investors and receive $6 million in seed funding “instantly, ” according to Pousaz — a clear sign of excess in venture dealmaking.

Checkout.com itself saw its valuation zoom nearly threefold to $40 billion in January after a new equity round. The firm generated revenue of $252.7 million and a pre-tax loss of $38.3 million in 2020, according to a company filing.

Asked what his company’s valuation would be today, Pousaz said: “Valuation is something for investors who care about entry point and exit point.”

“The multiples last year are not the same multiples than this year,” he added. “We can look at the public markets, the valuations are mostly half what they were last year.”

“But I would almost tell you that I don’t care at all because I care about where my revenue is going and that’s what matters,” he added.

Private tech company valuations are under immense pressure amid rising interest rates, high inflation and the prospect of a global economic downturn. The Fed and other central banks are raising rates and reversing pandemic-era monetary easing to stave off soaring inflation.

That’s led to a sharp pullback in high-growth tech stocks which has, in turn, impacted privately-held startups, which are raising money at reduced valuations in so-called “down rounds.” The likes of Stripe and Klarna have seen their valuations drop 28% and 85%, respectively, this year.

“What we’ve seen in the last few years was a cost of money that was 0,” Pousaz said. “That’s through history very rare. Now we have a cost of money that is high and going to keep going higher.”

Higher rates spell challenges for much of the market, but they represent a notable setback for tech firms that are losing money. Investors value companies based on the present value of future cash flow, and higher rates reduce the amount of that expected cash flow.

Pousaz said investors are yet to find a “floor” for determining how much the cost of capital will rise.

“I don’t think anyone knows where the floor is on the upper hand,” he said. “We need to reach the floor on the upper hand to then decide and start predicting what is the lower end, which is the long term residual cost of capital.”

“Most investors do valuations still to this day on DCF, discounted cash flow, and to do that you need to know what is the residual floor on the downside. Is it 2%, is it 4%? I wish I knew. I don’t.”

A common topic of conversation at Web Summit was the relentless wave of layoffs hitting major tech companies. Payments firm Stripe laid off 14% of its employees, or about 1,100 people. A week later, Facebook owner Meta slashed 11,000 jobs. And Amazon is reportedly set to let go 10,000 workers this week.

“I think every investor is trying to push this to their portfolio companies,” Tamas Kadar, CEO of fraud prevention startup Seon, told CNBC. “What they usually say is, if a company is not really growing, it’s stagnating, then try to optimize profitability, increase gross margin ratios and just try to just lengthen the runway.”

Venture deal activity has been declining, according to Kadar. VCs have “hired so many people,” he said, but many of them are “out there just talking and not really investing as much as they did before.”

Not all companies will make it through the looming economic crisis — some will fail, according to Par-Jorgen Parson, partner at VC firm Northzone. “We will see spectacular failures” of some highly valued unicorn companies in the months ahead, he told CNBC.

The years 2020 and 2021 saw eye-watering sums slosh around equities as investors took advantage of ample liquidity in the market. Tech was a key beneficiary thanks to societal shifts brought about by Covid-19, like working from home and increased digital adoption.

As a result, apps promising grocery delivery in under 30 minutes and fintech services letting consumers buy items with no upfront costs and virtually anything to do with crypto attracted hundreds of millions of dollars at multibillion-dollar valuations.

In a time when monetary stimulus is unwinding, those business models have been tested.

“An entire industry got ahead of its skis,” Parson said in an interview. “It was very much driven by hedge fund behaviour, where funds saw a sector that is growing, got exposure to that sector, and then bet on a number of companies with the expectation they will be the market leaders.”

“They pushed up the valuation like crazy. And the reason why it was possible to do that was because there were no other places to go with the money at the time.”

Maelle Gavet, CEO of pre-seed investment firm Techstars, agreed and said some later-stage companies were “not built to be sustainable at their current size.”

“A down round may not be always possible and, frankly, for some of them even a down round may not be a viable option for external investors,” she told CNBC.

“I do expect a certain number of late stage companies basically disappearing.”

JPMorgan Chase has hired a Singapore-based quantum-computing expert to be the bank’s global head for quantum communications and cryptography, according to a memo obtained by CNBC.

Charles Lim, an assistant professor at the National University of Singapore, will be focused on exploring next generation computing technology in secure communications, according to the memo from Marco Pistoia, who runs the bank’s global technology applied research group.

Lim is a “recognized worldwide leader” in the area of quantum-powered communications networks, according to Pistoia.

Hired from IBM in early 2020, Pistoia has built a team at JPMorgan focused on quantum computing and other nascent technologies. Unlike today’s computers, which store information as either zeros or ones, quantum computing hinges on quantum physics. Instead of being binary, qubits can simultaneously be a combination of both zero and one, as well as any value in between.

The futuristic technology, which involves keeping hardware at super-cold temperatures and is years away from commercial use, promises the ability to solve problems far beyond the reach of today’s traditional computers. Technology giants including Alphabet and IBM are racing toward building a reliable quantum computer, and financial firms including JPMorgan and Visa are exploring possible uses for it.

“New horizons are going to become possible, things we didn’t think would be possible before,” Pistoia said in a JPMorgan podcast interview.

In finance, machine-learning algorithms will improve to help fraud detection on transactions and other areas that involve “prohibitive complexity,” including portfolio optimization and options pricing, he said.

Drug development, materials science for batteries and other areas will be transformed by the dramatically advanced computing, he added.

But if and when the advanced computing technology becomes real, the encryption techniques that underpin the world’s communications and financial networks could immediately be rendered useless. That has spurred the study of next generation quantum-resistant communication networks, which is Lim’s area of expertise.

New forms of cryptography and secure messaging are needed ahead of quantum supremacy, or the point when quantum computers are able to perform calculations beyond the scope of traditional computers in any reasonable time frame, Pistoia said during the podcast.

That could happen by the end of the decade, he said.

The quantum advantage precedes that development and could happen as soon as two or three years from now, he said. At that point the new computers are more powerful and accurate than today’s versions but they are competitive.

“Even now that quantum computers are not yet that powerful, we don’t have so much time left,” Pistoia said in the podcast. That’s because bad actors are already preserving private communications to attempt to decrypt it later when the technology allows for it, he said.

Lim will “pursue both foundational and applied research in quantum information, focusing on innovative digital solutions that will enhance the security, efficiency, and robustness of financial and banking services,” Pistoia said in the memo.

Lim is a recipient of the National Research Foundation Fellowship in Singapore and won the National Young Scientist Award in 2019 for his work in quantum cryptography, said Pistoia.

Last year, Lim was asked to lead his country’s effort to create quantum-resistant digital solutions, and he has been involved in international efforts to standardize quantum-security techniques, Pistoia added.

Tesla CEO Elon Musk sent Twitter shares tumbling on Friday when he said he was going to put his $44 billion acquisition of the social network “on hold” while he researches the proportion of fake and spam accounts on the platform.

Though Musk later clarified that he remains committed to the deal, he continued to hammer on the issue of fake accounts. He wrote, on Twitter, that his team would do their own analysis and expressed doubt about the accuracy of numbers Twitter has reported in its most recent financial filings.

In its first-quarter earnings report this year, Twitter acknowledged there are a number of “false or spam accounts” on its platform, alongside legitimate monetizable daily active usage or users (mDAU). The company reported, “We have performed an internal review of a sample of accounts and estimate that the average of false or spam accounts during the first quarter of 2022 represented fewer than 5% of our mDAU during the quarter.”

Twitter also admitted to overstating user numbers by 1.4 million to 1.9 million users over the past 3 years. The company wrote, “In March of 2019, we launched a feature that allowed people to link multiple separate accounts together in order to conveniently switch between accounts,” Twitter disclosed. “An error was made at that time, such that actions taken via the primary account resulted in all linked accounts being counted as mDAU.”

While Musk may be justifiably curious, experts in social media, disinformation and statistical analysis say that his suggested approach to further analysis is woefully deficient.

Here’s what the SpaceX and Tesla CEO said he would do to determine how many spam, fake and duplicate accounts exist on Twitter:

“To find out, my team will do a random sample of 100 followers of @twitter. I invite others to repeat the same process and see what they discover.” He clarified his methodology in subsequent tweets, adding: “Pick any account with a lot of followers,” and “Ignore first 1000 followers, then pick every 10th. I’m open to better ideas.”

Musk also said, without providing evidence, that he picked 100 as the sample size number for his study because that’s the number Twitter uses to calculate the numbers in their earnings reports.

“Any sensible random sampling process is fine. If many people independently get similar results for % of fake/spam/duplicate accounts, that will be telling. I picked 100 as the sample size number, because that is what Twitter uses to calculate <5% fake/spam/duplicate.”

Twitter declined to comment when asked if his description of its methodology was accurate.

Facebook co-founder Dustin Moskovitz weighed-in on the issue via his own Twitter account, pointing out that Musk’s approach is not actually random, uses a too small sample, and leaves room for massive errors.

He wrote, “Also I feel like ‘doesn’t trust the Twitter team to help pull the sample’ is it’s own kind of red flag.”

BotSentinel founder and CEO Christopher Bouzy said in an interview with CNBC that analysis by his company indicates that 10% to 15% of accounts on Twitter are likely “inauthentic,” including fakes, spammers, scammers, nefarious bots, duplicates, and “single-purpose hate accounts” which typically target and harass individuals, along with others who spread disinformation on purpose.

BotSentinel, which is primarily supported through crowdfunding, independently analyzes and identifies inauthentic activity on Twitter using a mix of machine learning software and teams of human reviewers. The company monitors more than 2.5 million Twitter accounts today, primarily English-language users.

“I think Twitter is not realistically classifying ‘false and spam’ accounts,” Bouzy said.

He also warns that the number of inauthentic accounts can appear higher or lower in different corners of Twitter depending on topics being discussed. For example, more inauthentic accounts tweet about politics, cryptocurrency, climate change, and covid than those discussing non-controversial topics like kittens and origami, BotSentinel has found.

Carl T. Bergstrom, a University of Washington professor who co-wrote a book to help people understand data and avoid being taken in by false claims online, told CNBC that sampling one hundred followers of any single Twitter account should not serve as “due diligence” for making a $44 billion acquisition.

He said that a sample size of 100 is orders of magnitude smaller that the norm for social media researchers studying this sort of thing. The biggest issue Musk would face with this approach is known as selection bias.

Bergstrom wrote in a message to CNBC, “There’s no reason to believe that followers of the official Twitter account are a representative sample of accounts on the platform. Perhaps bots are less likely to follow this account to avoid detection. Perhaps they’re more likely to follow to seem legitimate. Who knows? But I just can’t fathom that Musk is doing anything other than trolling us with this silly sampling scheme.”

In a court filing out late Friday, shareholders who are suing Tesla and CEO Elon Musk over alleged securities fraud said they won part of a critical ruling in their class-action lawsuit.

The shareholders are suing Tesla over money they lost after Musk tweeted in 2018 that he was considering taking his electric vehicle company private at $420 per share and said he had funding secured to do so.

Tesla’s stock trading initially halted, then shares were highly volatile for weeks after the tweets. Musk later said that he had been in discussions with Saudi Arabia’s sovereign wealth fund and felt confident that funding would come through at his proposed price. A deal never materialized.

The Securities and Exchange Commission investigated and charged Musk with civil securities fraud as a result of those tweets. Tesla and Musk struck a revised settlement agreement in 2019 over those charges, but Musk is trying to terminate that agreement now.

Damages from the shareholders’ class-action lawsuit could amount to billions of dollars that would be paid by Musk and Tesla to those who are members of the class.

The shareholders’ attorneys said in the filing out Friday that Judge Edward M. Chen, who is presiding in this matter, had concluded that Musk acted with scienter — in other words, that he knowingly made false statements about having funding secured when he tweeted.

This information was revealed in a request the shareholders’ lawyers made for a temporary restraining order against Musk to stop him from making further public remarks about aspects of this case, as he did during a widely viewed appearance at the TED 2022 conference on April 14.

The request for the temporary restraining order alludes to an earlier ruling by Judge Chen that is currently under seal because it refers to evidence that Musk’s team regarded as confidential. “We anticipate the order will be published soon,” Adam Apton of Levi & Korsinsky, lead counsel for the class of Tesla shareholders, told CNBC by email.

At the TED conference on Thursday, Musk called financial regulators in the SEC’s San Francisco office “bastards.”

Musk also said, “The SEC knew that funding was secured but they pursued an active, public investigation nonetheless at the time. Tesla was in a precarious financial situation. And I was told by the banks that if I did not agree to settle with the SEC that they would, the banks would cease providing working capital and Tesla would go bankrupt immediately. So that’s like having a gun to your child’s head. I was forced to concede to the SEC unlawfully.”

It’s not clear why Musk felt he may have been unable to obtain working capital for Tesla, but confident he could muster the billions required to take the company private at the same time.

Musk is currently the richest person in the world on paper, and is trying to acquire Twitter, his social media platform of choice, and take it private for around $43 billion.

Musk’s attorney Alex Spiro, a partner at Quinn Emanuel Urquhart & Sullivan, said in a statement emailed to CNBC: “Nothing will ever change the truth which is that Elon Musk was considering taking Tesla private and could have – all that’s left some half decade later is random plaintiffs’ lawyers trying to make a buck and others trying to block that truth from coming to light all to the detriment of free speech.” 

Spiro gave the same statement to Bloomberg, which first reported on new developments in the shareholders’ class action.

A trial date is currently set for May 31, 2022, in a San Francisco federal court, but that could change.

Levi & Korsinsky’s Apton told CNBC, “We look forward to proving the rest of our case at trial and recovering damages on behalf of the class.”