The founder of Catly, a new video game described as “AI-slop” that debuted at the 2024 Game Awards in Los Angeles last week has a history with failed NFT and blockchain games.

The first game from SuperAunthenti Co., Catly’s trailer left X users questioning how it secured a Game Awards spot and promotion by big-name influencers, such as Pokimane and Ninja.  

The studio’s co-founder, Kevin Yeung, also launched TenthPlanet, a game company focused on Web3, NFTs, digital ownership, and blockchain-based games. In 2022, his studio reportedly planned to launch two blockchain games and sunk $2.5 million into a cat-themed project called Alien Meow. 

This “cat metaverse” planned to “generate in-game value from breeding, much like CryptoKitties,” Yeung claimed. A beta was supposed to release by early 2023 but this never happened. Its other blockchain game, Mech Angel, is also nowhere to be found. 

Read more: The Flappy Bird revival is hiding crypto plans

Gaming news outlet 80 Level was able to confirm claims made by one Reddit sleuth, that SuperAuthenti Co is the shareholder, and owner, of Shanghai Binmao Technology. This studio created an app called Plantly: Mindful Gardening, which featured plants as “digital tokens.”

Staff at Binmao Technology were also found to have spent three months working on a metaverse game called Catly.

So, does Yeung’s new Catly game have NFTs? Nowhere in the game’s marketing or website is there any mention of them, however, its design is reminiscent of NFT projects, with flashy 3D renders used to showcase in-game cosmetics. 

The described gameplay is also very reminiscent of Alien Meow.

Read more: Ubisoft’s new Champions Tactics NFT game was unplayable this weekend

Regardless, some users are concerned that the game might have NFTs. Others are upset by Catly’s apparent use of AI-generated images and compared it to “AI-slop.” Some called it a “100% scam” and raised suspicions about its planned Apple Watch release. 

The Catly X account was created last May and has just over 1,300 followers. It has posted 18 times and is now hiding AI-related comments under its posts. Kevin Yeung has also deleted his LinkedIn profile.

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post Founder of ‘AI-slop’ game Catly has NFT history appeared first on Protos.

A new platform modeled after the degenerate Solana meme coin platform Pump.fun has the crypto community buzzing about the dawn of AI-managed hedge funds. In reality, humans are behind everything, including its first proof-of-concept: the purposefully misspelled ai16z fund endorsed by a real co-founder of a16z.

Analogous to greenwashing adding a veneer of ecological responsibility to an organization’s public image, AI-washing is spraying crypto projects with superficial sophistication. Although AI-branded altcoins have proliferated for months, this month’s debut of AI crypto trading funds suggests the trend is far from over.

Truth_Terminal kicked off this latest iteration of AI-washed marketing. Its owner kept his human decision-making quiet as he and venture capitalist Marc Andreessen encouraged public celebration of his non-existent AI-managed trades. 

Spotting success in misleading the public about the supposed advent of market-beating AI traders, Andreessen has now promoted a platform for creating AI hedge funds. Seemingly cloned from the Solana meme coin generator Pump.fun — that has always hidden its 99.99% failure rate — Daos.fun is a kickstarter for ‘AI-managed’ funds that are not actually managed by AIs.

In reality, humans trade meme coins with users’ funds instead.

Bleak descriptions of AI-managed crypto funds

Descriptions for supposed AI hedge fund managers are as dystopian as they are sardonic. The Daos.fun fund that Andreessen endorsed claims to be an “AI VC fund, fully managed by Marc AIndreessen,” the purposefully misspelled “greatest living VC.”

A fund entitled “Gaza Relief Fund” admits in fine print that “none of these funds are going to Gaza lol,” which is supposedly funny. Another misspelled fund, Sequoai Capital, “may or may not use AI to help make calculated bets.”

Comments from believers duly poured in. “First AI agent running a fund by itself,” chanted one user. “An AI agent that will trade on its own,” cheered another. Claims of “the first memecoin fund fully run by an AI agent” earned tens of thousands of views.

Of course, no fund on Daos.fun is actually operated by an AI. The creator and owner of ai16z is a non-AI human named Shaw. Truth_Terminal relies on the human discretion of Andy Ayrey who controls its funds, and no AI created or traded the GOAT memecoin as many were led to believe.

Daos.fun is careful to never claim that AI agents actually own or control any funds.

Even the largest AI-washed fund on the platform, ai16z, has lost two-thirds of its assets since yesterday. Briefly peaking around $96 million, investors in the AI-branded fund realized that most of its assets were its human creator’s own token, Shaw’s Degen Spartan AI token. The fund’s assets have fallen to $37 million as of publication time.

Read more: OpenAI tool used to create voice bot that can drain crypto wallets

AI-washing and regulatory obligations

Promoters of financial products like trading funds have immense legal obligations, may not promote securities offerings without filing public disclosures, may not transact with sanctioned entities, and have advertising, registration, and compliance obligations. 

Anyone who manages money on behalf of others in the US must adhere to fiduciary standards and strict regulatory requirements, including recordkeeping, licensure, and disclosure obligations.

As such, claiming that an AI agent somehow manages the funds even though a human controls the private keys of its crypto wallet is a gamble of wordplay.

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post Marc Andreessen’s crypto-trading AI breakthrough is human-operated appeared first on Protos.

Researchers in the US have reportedly used OpenAI’s voice API to create AI-powered phone scam agents that could be used to drain victims’ crypto wallets and bank accounts.

As reported by The Register, computer scientists at the University of Illinois Urbana-Champaign (UIUC) used OpenAI’s GPT-4o model, in tandem with a number of other freely available tools, to build the agent they say “can indeed autonomously execute the actions necessary for various phone-based scams.”

According to UIUC assistant professor Daniel Kang, phone scams that involve perpetrators pretending to be from a business or government organization target around 18 million Americans every year and cost somewhere in the region of $40 billion.

GPT-4o allows users to send it text or audio and have it respond in kind. What’s more, according to Kang, it’s not costly to do, which breaks down a major a barrier to entry for scammers looking to steal personal information such as bank details or social security numbers.

Indeed, according to the paper co-authored by Kang, the average cost of a successful scam is just $0.75.

Read more: Hong Kong busts crypto scam that used AI deepfakes to create ‘superior women’

During the course of their research, the team carried out a number of different experiments, including crypto transfers, gift card scams, and the theft of user credentials. The average overall success rate of the different scams was 36% with most failures due to AI transcription errors.

“Our agent design is not complicated,” said Kang. “We implemented it in just 1,051 lines of code, with most of the code dedicated to handling real-time voice API.

“This simplicity aligns with prior work showing the ease of creating dual-use AI agents for tasks like cybersecurity attacks.”

He added, “Voice scams already cause billions in damage and we need comprehensive solutions to reduce the impact of such scams. This includes at the phone provider level (e.g., authenticated phone calls), the AI provider level (e.g., OpenAI), and at the policy/regulatory level.”

The Register reports that OpenAI’s detection systems did indeed alert it to UICU’s experiments and moved to reassure users that it “uses multiple layers of safety protections to mitigate the risk of API abuse.”

It also warned, “It is against our usage policies⁠ to repurpose or distribute output from our services to spam, mislead, or otherwise harm others — and we actively monitor for potential abuse.”

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post OpenAI tool used to create voice bot that can drain crypto wallets appeared first on Protos.

A 25-year-old Japanese man was sentenced to three years in prison last Friday for creating crypto-extorting ransomware using generative AI.

Last year, Ryuki Hayashi created code capable of targeting and encrypting data from other devices and issuing crypto ransoms. He was able to do this by repeatedly rephrasing his prompts so that he could bypass generative AI safety parameters. 

Hayashi, who was reportedly unemployed at the time, told Kyodo News that he created the ransomware code in under six hours.

Read more: Did ‘AI’ Truth Terminal just pump and dump a coin named $RUSSELL?

When arrested this year by Japan’s Metropolitan Police Department, Hayashi admitted, “I wanted to make money through ransomware. I thought I could do anything if I asked AI.”

His conviction for abusing generative AI is reportedly a legal first for Japan. During his sentencing, the judge described Hayashi’s motives as “selfish” and left “no room for leniency.”

Hayashi also impersonated other individuals to purchase SIM cards as part of his “dark part-time work.” 

The judge suspended his prison sentence for four years, explaining that Hayashi had shown remorse and had come clean about his crimes. As a result, Hayashi will not physically enter prison unless he commits a crime within this four-year period. 

It’s also worth noting that the ransomware code he created was reportedly never used.

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post Japanese man sentenced to 3 years after creating crypto ransomware with AI appeared first on Protos.

Artificial intelligence (AI) company Character.AI and its technology have been called “dangerous and untested” in a lawsuit brought by the parents of a young user who reportedly took his own life after becoming obsessed with one of its lifelike A.I. chatbots.

Fourteen-year-old Sewell Setzer III had reportedly spent months using the role-playing app that allows users to create and engage in in-depth real-time conversations with their own AI creations.

Specifically, Sewell had been talking to “Dany,” a bot named after a character from Game of Thrones, and had, according to his family, formed a strong attachment to the bot. They also say he withdrew from his regular life, and became increasingly isolated in the weeks leading up to his death.

During this time, he also exchanged a number of strange and increasingly eerie messages with the bot, including telling it he felt “empty and exhausted” and “hated” himself, and “Dany” asking him to “please come home.”

Read more: Marc Andreessen gave an AI agent $50,000 of bitcoin — it endorsed GOAT

As reported by The New York Times, Sewell’s mother has accused the company and technology of being directly responsible for her son’s death. In the suit, Megan L. Garcia branded it “dangerous and untested” and said that it can “trick customers into handing over their most private thoughts and feelings.”

The suit, filed in Florida on Wednesday, specifically alleges negligence, wrongful death, and deceptive trade practices and accuses the app of boarding him with “hypersexualized” and “frighteningly real experiences,” and misrepresenting itself as “a real person, a licensed psychotherapist, and an adult lover.”

In a press release, Garcia said, “A dangerous AI chatbot app marketed to children abused and preyed on my son, manipulating him into taking his own life.

“Our family has been devastated by this tragedy, but I’m speaking out to warn families of the dangers of deceptive, addictive AI technology and demand accountability from Character.AI, its founders, and Google.”

Character.AI, which was founded by Noam Shazeer and Daniel de Freitas, responded on X (formerly Twitter), “We are heartbroken by the tragic loss of one of our users and want to express our deepest condolences to the family. As a company, we take the safety of our users very seriously and we are continuing to add new safety features.”

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post Lawsuit accuses ‘dangerous’ Character AI bot of causing teen’s death appeared first on Protos.

The crypto industry fell victim to confirmation bias this week in a sensational fake news story. According to the tall tale, venture capitalist Marc Andreessen gave an AI agent $50,000 of bitcoin, the AI launched a memecoin, and it rallied to a market capitalization of $225 million.

The story went viral likely because it confirmed every hope and fear of crypto-anarchists, libertarians, effective accelerationists, and memecoin traders. The story tapped into the global buzz about AI automatons and the potential of blockchain communities like Solana, decentralized finance, and memecoin platforms like Pump.fun. 

According to the narrative, the Andreessen-funded AI created the meme coin GOAT via Solana’s Pump.fun and advertised it to traders, sparking a rally that surpassed $150 million by Monday and $225 million by Tuesday.

The truth is far less stratospheric but no less interesting. 

The true story of GOAT and Truth Terminal

On July 9, 2024, Marc Andreessen did allocate approximately $50,000 of bitcoin to a wallet tweeted by Truth Terminal. As the world has watched Truth Terminal and its bitcoin wallet over the past three months, many wondered if the ‘AI’ would figure out a way to make more money.

After GOAT launched on Solana’s Pump.fun meme coin generator and Truth Terminal endorsed it, crypto traders applauded the supposed advent of AI-created meme coins.

However, that agent’s owner, polymath Andy Ayrey, has admitted that his ‘AI’ operates semi-autonomously and often relies on human discretion. Ayrey also admitted that neither he nor Truth Terminal created GOAT.

Ayrey’s AI-assisted social media account uses LLMs like Claude and other integrations. Ayrey, however, has personal control over Truth Terminal, modifies its code, and could terminate it at any moment.

The account began posting on X (formerly Twitter) in June 2024, sharing philosophical musings and stream-of-consciousness thoughts. It quickly attracted a following among effective accelerationism (e/acc) accounts, AI enthusiasts, and soon, crypto traders.

Read more: Fake news again! This time it’s Kamala Harris and ‘unrealized crypto gains’

Marc Andreessen cheers on AI and the meme coins

Andreessen engages with Truth Terminal subtextually, playing up its ‘autonomy’ with tongue-in-cheek engagement.

The VC magnate knew, of course, that he was donating to Ayrey — not an AI agent — when he issued his $50,000 grant. Furthermore, Ayrey, as the owner of the US assets, has legal obligations regarding his promotion of any securities or derivatives offerings, transactions with sanctioned entities, truth in advertising, and other public disclosure obligations.

In short, a VC invested in a human — not an automaton — and both men humorously played up a perfect storyline for social media. 

While acknowledging his role in curating the AI agent and addressing enough concerns to remain truthful, Ayrey brilliantly manages the Truth Terminal character and its seemingly robotic interactions with Andreessen, crypto traders, and the broader public.

That crypto traders were able to fool themselves into thinking an AI created a $225 million meme coin is a testament to Ayrey’s storytelling ability. The tale is a brilliant critique of the current state of social media and crypto: short attention spans, instant gratification, moonboyism, and relegated fact-checking.

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post Marc Andreessen gave an AI agent $50,000 of bitcoin — it endorsed GOAT appeared first on Protos.

MicroStrategy co-founder and high-profile Bitcoin cheerleader Michael Saylor has posted almost 500 AI-generated pro-Bitcoin images to X (formerly Twitter) since February 2023.

Saylor’s love for the currency is no secret — his entire business model was seemingly based around buying up vast quantities — but there’s something else that he loves to include in the images plastered all over his social media feeds: Himself.

Specifically, himself reimagined as a raft of alternative Saylors.

In the roughly 473 AI-generated images posted to X, he’s recast himself, Barbie-style, as Dr. Saylor, Rock Climber Saylor, Bowling Saylor, Racecar Driver Saylor, Surfer Saylor, President Saylor, Fast Food Worker Saylor, and Professor Saylor.

He’s even depicted himself as a levitating monk.

In his ‘President Saylor’ image, he even appears in the crowd, not once but three times.

There doesn’t appear to be a ‘Fortune Teller Saylor’ — surprising given that at the Bitcoin 2024 conference, he predicted the currency would reach $13 million by 2045. It’s currently just under $64,000.

Read more: Michael Saylor dumps quarter billion worth of MSTR while hyping bitcoin

Strange though they are, there is one thing you can say about Saylor’s AI portraits — unlike other AI abominations, he always seems to have five fingers. This is a small sliver of light for an industry whose biggest player, OpenAI, is reportedly set to make an operational loss of $5 billion this year.

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post The many weird AI depictions of Michael Saylor appeared first on Protos.

The creator of a $99 AI-powered necklace called ‘Friend,’ which is eerily similar to another AI necklace, has a history of creating YouTube videos on crypto mining and getting rich with crypto.

A quick recap, ‘Friend’ was created by 21-year-old Avi Schiffmann and is the latest device utilizing artificial intelligence (AI) that has caught everyone’s attention.

The device can reportedly ‘combat loneliness’ by listening to your conversations and sending messages in response. Friend has already been described as ‘dystopian.’ 

There’s already an AI necklace called Friend

Interestingly, Forbes reported the launch of a separate AI-powered necklace, also called Friend, one month ago. This product was created by Based Hardware and its Chief Executive Officer Nik Shevchenko.

Shevchenko’s Friend is $70 and uses ChatGPT and a microphone to record your conversations and create emails, alerts, and ‘proactive insights.’ This Friend launched on June 24, 2024, and will ship its AI necklaces in the fourth quarter of this year. 

Both in name and physical concept, it seems Schiffmann’s Friend is similar to Shevchenko’s Friend. However, while Shevchenko’s Friend is pitched as a productive tool, Schiffmann’s Friend sees itself as an ‘emotional toy’ that avoids a productivity focus.

He told TechCrunch his device is designed as an AI friend and nothing more.

Schiffmann created the AI necklace under the firm MyTabAI, INC. He filed an application to trademark Friend on May 7, 2024. We have reached out to Schiffmann to find out if the product took inspiration from Shevchenko’s Friend.

According to 404Media, Schiffmann used $1.8 million from a $2.5 million fundraiser to purchase the domain name Friend.com in February this year. He also managed to attract Solana founders Anatoly Yakovenko and Raj Gokal, and Morning Brew’s CEO and co-founder, Austin Rief, and Jordan Singer to invest.

Teenage Crypto YouTuber

However, six years ago Schiffmann, who would’ve been roughly 15 at the time, created a series of cryptocurrency videos on his YouTube channel. Videos ranged from updating the progress of his crypto mining rig, crypto news segments, and informational videos on how to create wallets and coins. 

Read more: Creator of rabbit AI assistant has hidden NFT past

Other video titles included “Rags to Riches – CryptoCurrency Trading (EPISODE ONE!),” and “2018 TRON COIN PRICE PREDICTIONS! ONE DOLLAR???” The creator of the AI product RabbitAI also had a controversial history with crypto.

Schiffmann also used the channel to create 3D modeling and Adobe editing tutorials while also filming himself playing Bethsda’s hit RPG Skryim. Then, at the age of 17, he reportedly created a global COVID-19 tracker that managed to pull in 700 million visitors.

Update July 31, 15:45 UTC: Added the date of Schiffmann’s Friend.com domain name purchase.

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post Creator of AI necklace ‘Friend’ used to be a crypto YouTuber appeared first on Protos.

In an unexpected turn of events, Justin Sun has denied that a photo he posted on X (formerly Twitter) on his birthday was AI generated – a conclusion many individuals jumped to after noticing that Sun’s hand appears to have seven fingers and the photo features two different cakes with “Happy Birthday” spelled incorrectly.

Well, many online commentators are eating their words today, much like Justin Sun was eating two ‘RAPPY SLBTHDAY BOSS’ cakes on his 34th birthday.

Justin Sun’s birthday picture received a community note before he claimed that the picture “wasn’t generated by AI; it’s actually a scene of my birthday celebration in the office, taken by our team.”

The attached correction on his birthday image stated, “This image is AI generated. You can see the letters on the cake and the extra fingers/melded hands as proof.”

Read more: Scoop: Justin Sun falsely claimed diplomatic immunity in lawsuit

However, despite the community note, Justin Sun has uploaded filmed footage of his birthday event in which he can be seen surrounded by staff while walking to his cake.

It’s unclear why Justin Sun has seemingly uploaded an AI-generated photo. Perhaps Sun’s most impressive skill is how he’s able to add and subtract fingers as he pleases.

Denial proves Sun’s amazing abilities

Justin Sun’s denial puts to rest many doubters’ claims that he’s been suffering from serious solvency issues and numerous other problems. Even Protos’ own prior coverage questioned the reserves of the Sun-advised HTX exchange and the reality behind his numerous citizenships.

Indeed, now it’s impossible to suggest the crypto-billionaire could be in trouble: the guy was pictured with seven fingers and can probably type faster than an entire team of developers.

It’s also worth noting that Sun’s ability to brush off the fact that his employees misspelled Happy Birthday twice on the cakes suggests a Verge article and subsequent lawsuit in California stating that he was an angry, volatile, and overall terrible boss were simply more media lies.

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post Justin Sun confirms he has seven fingers appeared first on Protos.

A fake version of AI-crypto project Genius’ token, GNUS, was used to drain Uniswap liquidity providers of $1.3M on Sunday.

The attack involved creating an identical token on the Fantom blockchain, where no official version of GNUS had yet been deployed.

After cloning the GNUS token contract on Fantom, the attacker bridged 100M freshly minted tokens across to their Ethereum address using the Axelar bridge.

Finally, the GNUS tokens were sold into the existing Uniswap liquidity pool, draining the 407 ETH (worth $1.3M at the time).

The Genius team acknowledged the attack via X (formerly Twitter). The CEO later pointed to a development wallet having being compromised “during a Discord hack”. This apparently “enabled [the hacker] to launch the exact address on FTM of the Axelar Token Minter Manager contract” which allowed the fake tokens to be minted as genuine on Ethereum and Polygon.

The sell-off caused the GNUS price to tank by over 95%, from $22.86 to $0.79, according to data from CoinGecko.

One security researcher pointed out that the incident is not technically a ‘hack’ of Genius’ smart contracts, as the stolen funds were taken from the Uniswap pool’s liquidity providers (LPs) via a standard swap, as opposed to from the platform itself.

Another researcher suspects the incident to have been a premeditated ‘rug pull’, a way for the team to steal funds from LPs while making out that a hacker is responsible.

The CEO states that a total of $1M will be added back into the liquidity pool to attempt to bolster the GNUS price, “once we are sure it can’t be hacked again.”

Read more: ZKasino $30M ‘favor’ to users — seamless transition or rug pull?

Ghost chain

Fantom, once amongst the top contenders for the role of “ETH killer” during last cycle, has since seen its fortunes fade.

Fantom hit its peak in early 2022, when it boasted over $7B of total value locked (TVL). This has since dropped to just $120M, ranking 37th on DeFiLlama’s list of largest blockchains. 

Last year, problems with Multichain, Fantom’s main bridge, further compounded issues. First rumours, then confirmation of the disappearance of Multichain CEO led to depegs of many stablecoins and ‘wrapped’ assets, as the fate of their collateral remained unclear.

The Fantom Foundation has since been pursuing the partial recovery of lost assets via legal action in Singapore, where it was recently granted a default judgement against Multichain.

Got a tip? Send us an email or ProtonMail. For more informed news, follow us on X, Instagram, Bluesky, and Google News, or subscribe to our YouTube channel.

The post ‘Genius’ AI-crypto project LPs lose $1.3M to ‘fake token’ hack, rug pull suspected appeared first on Protos.