Ten countries including favorite Finland and contentious competitor Israel won places Tuesday in the Eurovision Song Contest final, while five nations were sent home after the first day of competition in the pop music extravaganza. But divisions are clouding the contest’s 70th anniversary edition, with five countries — Spain, Ireland, the Netherlands, Slovenia and Iceland — boycotting to protest Israel’s inclusion.
Samsung’s rumored Galaxy Z Fold 8 Wide may not offer top-tier cameras, with leaked specs pointing to a 50MP main shooter.
Since 1946, the Festival de Cannes (aka the Cannes Film Festival) in France has been a beacon of cinematic excellence and cultural exchange.
For those who love the Academy Awards, films such as Parasite and Anora debuted here first before taking home an Oscar. This year promises to continue this worthy legacy despite fewer American entries than normal. Here’s everything you need to know as the festivities kick off this week.
In July 1938, a Nazi propaganda film helped inspire Philippe Erlanger to create a new film festival. He was one of many who were displeased that Leni Riefenstahl’s Olympia and Goffredo Alessandrini’s Luciano Serra, Pilot took home the Mussolini Cup at the Venice Mostra due to pressure from Hitler.
Erlanger endeavored to create a fair competition based on merit, not politics. The first event was supposed to happen on September 1, 1939, but Germany invaded Poland and World War II was officially declared two days later.
The Cannes Film Festival was put on the back burner until peace could be obtained. The first festival would have included films such as The Wizard of Oz by Victor Fleming and Union Pacific by Cecil B. DeMille.
Six years later, in 1946, 19 countries finally participated, and Georges Huisman led the international jury. Every nation was awarded a Grand Prix as the world continued to heal from war. America’s entry was Billy Wilder’s The Lost Weekend.
Fast-forward to 2026, and 2,541 feature films were submitted for consideration. Twenty-two will be up for the coveted Palme d’Or.
South Korean filmmaker Park Chan-wook will preside over the jury that will decide the victor.
Other members include director Chloé Zhao, actress Demi Moore, and actor Stellan Skarsgård.
There are only two American entries in the competition category of films this year.
James Gray’s Paper Tiger was added after the initial announcement. This crime drama tells the story of the Pearl brothers in 1980s Queens. While trying to pursue the American dream, these men get involved with the Russian mafia instead.
The film will bring the Hollywood A-listers to the French Riviera as its stars Scarlett Johansson, Adam Driver, and Miles Teller.
The other American entry is Ira Sachs’s The Man I Love, starring Rami Malek. It is also set in 1980s New York. The plot centers around Jimmy George, an actor with a life-threatening illness, as he tackles his final role.
Ryusuke Hamaguchi’s All of a Sudden is the Japanese filmmaker’s first French movie. Fans are thrilled to see what this exciting voice has to say after his 2021 critically acclaimed flick Drive My Car. This was the first Japanese movie to be nominated for a Best Picture Academy Award.
You can find a full list of official sections on the festival’s website.
Beyond films, individuals will also receive awards.
Both director and writer Peter Jackson and multi-hyphenate Barbra Streisand will take home Honorary Palme d’Ors. This is essentially a lifetime achievement award for their impressive resumes.
Jackson is best known for his work on The Lord of the Rings and Hobbit trilogies. Streisand’s career covers everything from Hello, Dolly! to Meet the Fockers, and that’s not counting her time behind the camera directing and producing.
The Cannes Film Festival runs May 12 through the 23.
US President Donald Trump arrived in Beijing on Wednesday for his first visit to China since 2017. The stakes are high for the two-day summit, which analysts suggest could mark a tipping point in the relationship between the superpowers – likely in Chinese President Xi Jinping’s favour.
Researchers in South Korea have developed a ring-based sign language translator that works wirelessly and can recognize ASL and ISL words with roughly 88% accuracy.
Oops, it happened again. A celebrity was asked what they think about artificial intelligence and, after sharing their reflections, received intense blowback on social media.
The latest such case is Demi Moore, who is currently serving on the jury for the Cannes Film Festival. At a May 12 press conference meant to introduce the broader film event, Moore was asked by a journalist about AI, its impact on Hollywood, and potential regulation.
“I always feel ‘againstness’ breeds ‘againstness.’ AI is here,” Moore responded, clearly thinking on the spot. Rather than fight a “losing” battle, Moore suggested that artists figure out how to “work with” the technology. This, she opined, would be a far more productive path forward. The Substance star then proceeded to suggest that there is probably not enough being done in terms of regulating the technology, before concluding with one final and trite, though seemingly heartfelt, salvo.
“The truth is: There really isn’t anything to fear because what [AI] can never replace is what true art comes from, which is not the physical. It comes from the soul,” she asserted. “It comes from the spirit of each and every one of us sitting here . . . to each and every one of us that creates every day. And that they can never re-create through something that’s technical.”
Moore has since been pilloried in some corners of the internet. She’s facing both fair criticism and a bevy of offensive insults, many of which dismiss her as a pro-AI shill or, perhaps worse, a pro-AI dunce.
Moore joins a growing number of celebrities who have either volunteered to comment on, or been asked about, AI, and subsequently been sorted into camps of support and opposition.
On one side are skeptics like Guillermo del Toro, who would “rather die” than use generative AI, and Nicolas Cage, who is a “big believer in not letting robots dream for us.”
On the other are more accommodating voices like Sandra Bullock, who says AI should be used in a “constructive” way, and Reese Witherspoon, who, quite inartfully and with the verbiage of a sponsored-content hack, encouraged women to get in the game and use the tech. These statements all tend to come with attendant cheers or barbs from online fans eager to police any positive statements about the technology.
This new micro-trend of celebrity AI takes—and AI takedowns—comes as Hollywood looks to position itself in relation to a technology that stands to rapidly transform cinematic production on the whole. Through automation, AI could, critics argue, threaten jobs, further abuse intellectual property, cheapen the process of art-making, and fuel the influence of Silicon Valley firms over creative industries.
Of course, there’s also a flip side: AI advocates say that while the arrival of these new platforms does challenge a traditional business model, they also lower the barrier to entry and constitute a new way to democratize the creation of art. Think about it: Now anyone with access to a few powerful models can produce high-quality animations, even if they don’t have a multimillion-dollar film budget or fancy studio.
But the problem with forcing anyone, including celebrities, into pro- and anti-AI camps is that AI is already here. Artificial intelligence also continues to be a wildly vague term that can refer to anything from machine-learning algorithms used to catch typos in scripts and assist in video editing to extremely impressive visuals rendered with just a few prompts to a large language model. In the jumble of online discourse, all of these phenomena are swept together into pro-AI or anti-AI contingents.
Yes, celebrities are making all sorts of cringey comments on AI, but lambasting them for acknowledging the technology is here, likely already endemic, and even comes with some compelling use cases isn’t progressing the conversation. AI is currently shaping our digital and material lives in ways that are useful and exciting and noxious and terrifying, often through mechanisms that are mostly beyond the consumptive or creative purview of any one person.
What might be more important is pushing people to think specifically about what they mean when they talk about AI and more critically about the ways AI is influencing the distribution of power, wealth, and even creativity. Dealing with AI requires looking for a more equitable vision of these tools, rather than polarizing ourselves as pro or against a technological category that remains extremely poorly defined.
Consider the thoughts expressed by Paul Laverty, a screenwriter and lawyer also serving on the Cannes Film Festival jury, in a follow-up to Moore.
“I think we have to look at the first thing and see who owns it, because they decide on the algorithms that affect our lives in the deepest way,” Laverty said. “People are beginning to realize that we should not let these tech bros—billionaires who are mostly right-wing libertarians—dictate how we live our lives. What’s the effect to be [on] workers, beyond artists, ordinary workers and society and our children?”
Which is to say: Maybe we ought to spend less time policing celebrity AI takes and more time interrogating the people building the systems themselves.
Kelp DAO and Aave say the rsETH crisis is ending, with the exploit-related burn on Arbitrum completed and a 117,132 rsETH refill now underway.
The April 18 incident was a bridge-and-oracle style attack, where forged data led the system to believe rsETH had been burned when it had not.
Kelp is hardening its security model by adding more verification parties, raising confirmation thresholds, and ending risky L2-to-L2 routes.
Today, we are witnessing one of the most sophisticated recovery efforts in the history of the industry. Following a harrowing month of de-pegs and bad debt scares, Kelp DAO and Aave have officially signaled the end of the “rsETH Crisis.”
As of May 13, 2026, the recovery is in full swing. The “exploit-burn” on Arbitrum is complete, the refill of 117,132 rsETH has begun, and the security architecture of the most popular liquid restaking token (LRT) has been fundamentally rebuilt from the ground up. This isn’t just a technical patch; it’s a masterclass in ecosystem resilience.
To understand the recovery, we have to look back at the chaos of April 18, 2026. This wasn’t a standard smart contract bug or a simple key leak. According to a comprehensive post-mortem by Chainalysis, Kelp DAO was the victim of a high-precision RPC poisoning attack orchestrated by the North Korean Lazarus Group.
The target was the LayerZero Omnichain Fungible Token (OFT) adapter. The attackers compromised the downstream RPC nodes that the LayerZero verifiers relied on to observe the “source” chain (in this case, Uniswap’s Unichain L2). By feeding forged data to a single-verifier configuration, the attackers tricked the bridge into believing that 116,500 rsETH had been burned on Unichain, when in reality, the supply was still there. The bridge, acting on the “verified” message, released the equivalent amount of rsETH on the Ethereum mainnet directly into the hacker’s lap.
This was an “observation-layer” exploit. It exposed a critical vulnerability in DeFi’s infrastructure: even a perfect smart contract is only as secure as the data feed it trusts. The fallout was immediate. The stolen rsETH was used as collateral on Aave v3 and Compound to borrow WETH, creating nearly $300 million in bad debt and causing the rsETH peg to drop as low as $2,800 while ETH traded at $3,500.
The update shared by Kelp DAO today marks the transition from “damage control” to “restoration.” The recovery involves a highly coordinated movement of assets between the Aave Recovery Guardian and the Kelp DAO Recovery Safe.
Over the next 14 days, a total of 117,132 rsETH will be progressively refilled into the LayerZero OFT adapter on Ethereum mainnet. This refill ensures that every single rsETH token circulating across the 20+ supported Layer 2s is once again backed 1:1 by real collateral in the mainnet escrow.
”rsETH on Mainnet and L2s remains fully backed at all times during this transition,” the team confirmed.
Crucially, the first tranche of this refill to the LayerZero OFT adapter is the “green light” for users. Kelp DAO intends to unpause withdrawals within 24 hours of this initial deposit. Once the contracts are unpaused, all standard operations, including redemptions, claims, and bridging, will resume as usual. For the thousands of users who have had their capital sidelined for weeks, this is the light at the end of the tunnel.
One of the most complex pieces of the recovery puzzle was dealing with the rsETH still held by the exploiter on Arbitrum. Because the attacker had posted the stolen tokens as collateral, they effectively had a “claim” on the system that threatened the recovery’s integrity.
Working closely with the Arbitrum Security Council and Aave governance, the recovery coalition managed to isolate and burn the exploiter’s rsETH holdings on Arbitrum. This “surgical removal” of the illicitly minted tokens was the prerequisite for the refill. By burning the hacker’s shadow supply, the team ensured that the new ETH being injected into the system actually backs legitimate user tokens, rather than providing an exit for the Lazarus Group.
Kelp DAO isn’t just refilling the coffers; they are building a fortress. The protocol recently completed a rigorous “security hardening pass” audited by BailSec. The goal was to eliminate the “single point of failure” that allowed the April exploit to happen.
This shift toward multi-source verification is a direct response to the RPC poisoning method used by Lazarus. By requiring consensus from four different organizations, Kelp DAO has ensured that an attacker would need to compromise the infrastructure of multiple independent firms simultaneously—a feat that is orders of magnitude more difficult than attacking a single node
Perhaps the most significant long-term development is Kelp DAO’s decision tomigrate away from LayerZero and toward Chainlink CCIP. This move reflects a growing rift between Kelp and LayerZero regarding responsibility for the April 18 incident.
While LayerZero maintains that the exploit was the result of Kelp’s “misconfigured” 1-of-1 DVN setup, Kelp DAO argues that the default settings and lack of timely infrastructure warnings were the root cause. By choosing Chainlink’s Cross-Chain Interoperability Protocol (CCIP), Kelp DAO is opting for a model that requires consensus from 16 independent node operators.
The move to the Chainlink Cross-Chain Token (CCT) standard is expected to be completed in the coming months. This transition marks a broader industry trend in 2026: as cross-chain volumes surge, “convenience” is being sacrificed for “verifiable security.”
The recovery of Kelp DAO’s rsETH is a testament to the maturation of the decentralized financial system. A year ago, a $292 million exploit might have triggered a catastrophic contagion that wiped out secondary lending markets. In 2026, we saw Aave, Mantle, and DeFi United step in within hours to form a “Recovery Guardian” coalition.
From Stani Kulechov’s personal 5,000 ETH pledge to the Arbitrum governance vote that paved the way for the exploiter’s burn, the recovery proves that “community” is more than just a buzzword in DeFi—it is a defensive layer.
As withdrawals unpause and rsETH operations return to normal, the takeaway for the rest of the market is clear: Infrastructure is the new battleground. In the Alpenglow and CCIP era, the protocols that survive won’t be the ones that ignore risk, but the ones that build systems resilient enough to recover from it.
A Dutch ornithologist may be “patient zero” in the new hantavirus outbreak – a theory currently favored by the World Health Organisation. The 70 year old was the first person to die aboard the cruise ship MV Hondius. But how and where did he come into contact with the virus?
Four Ethereum whales accumulated 7788 ETH worth approximately $17.67 million.
Samsung’s next foldables could debut Gemini Intelligence, giving the Galaxy Z Fold 8 and Z Flip 8 a more capable AI assistant that handles linked tasks across apps from a single request.
