Castlevania: Legacy of Darkness, the director's cut version of Castlevania for the Nintendo 64, is a classic today despite issues like blurry graphics and wonky gameplay. You'd think a title designed in the last century (well, 1999) would have no new surprises, but there's a twist to the story. A new Konami Code has been found and it's interesting enough that Castlevania fans may want to give it a fresh look. 

The trick was first revealed by Moises and LiquidCat on the Castlevania 64 Discord server, as shared by YouTuber JupiterClimb. To unlock it, you just need to double the inputs of the original Konami Code (press ↑ ↑ ↓ ↓ ← → ← → B A, twice), and all the characters from Castlevania: Legacy of Darkness — Cornell, Reinhardt, Carrie and Henry — are instantly unlocked. 

It also opens up all their alternate costumes, along with the game's hard mode. Without the code, you have to win the game (twice!!) to unlock all four characters. 

Along with the main code, Moises and LiquidCat also dug up codes to fill all inventories, regardless of character being played. The PAL and JPN version also lets players max out their jewels and weapon power. 

JupiterClimb speculates that the code was a developer shortcut accidentally left in the game, as has happened before with Konami titles. In any case, it's a fun surprise that breathes some new life into the 25-year-old vampire game. 

Rivian will give you up to around $5,470 in discount if you trade in an eligible gas-powered truck or SUV when you purchase or lease a qualifying R1 electric vehicle package in the US and Canada. As an "Electric Upgrade Offer" for Earth Day, Rivian said it will accept 2018 or newer Ford F-150, Explorer, Expedition and Bronco (excluding Bronco Sport) vehicles, as well as 2018 or newer Toyota Tacoma, Tundra, Highlander and 4Runner vehicles for trade in. You can also trade in a Jeep Grand Cherokee, Wrangler or Gladiator from the same model years. Rivian will take a 2018 or newer Audi Q5, Q7 and Q8 and a BMW X3, X5 and X7, as well. 

As you can see, some of those models are incredibly popular gas vehicles, like the Ford F-150, as the company is likely hoping to appeal to a wide range of people who may be considering switching to electric. Of course, the amount you get will depend on your vehicle and its condition, which means you could get more if you sell it yourself. In addition, the discount will only apply to specific R1T truck and R1S SUV packs at amounts that range from CAD$1,000 ($730) to CAD$7,500 ($5,470). You'll also still have to put in a $1,000 non-refundable deposit to reserve the configuration you choose, and you must be able to accept a delivery between April 22 and June 30.

As TechCrunch notes, Rivian launched the promo at a time when there's lower demand for electric vehicles, especially for more expensive premium models. Other automakers recently introduced discounts of their own — Tesla, for instance, shaved $2,000 off the starting prices of the Model Y, Model X and Model S. It's also ending its referral program on April 30 and is making its Full-Self Driving software $4,000 cheaper. If you do trade in an eligible gas vehicle to buy an R1, you'll also be able to charge your new EV at all Rivian Adventure Network sites for free for one year.

Meta announced it’s opening up the Quest’s operating system to third-party companies, allowing them to build headsets of their own. The Quest OS is being rebranded to Meta Horizon OS and already has two companies interested.

ASUS’ Republic of Gamers (ROG) brand is working on new “performance gaming” headsets, while Lenovo’s focus is on devices for “productivity, learning and entertainment.” However, most intriguingly, perhaps, Meta says it’s also working on a limited-edition Xbox “inspired” Quest headset. (Microsoft and Meta also worked together recently to bring Xbox cloud gaming to the Quest.) While this could just be a reskinned Quest 3, this collaboration could lead to future headsets made entirely for Microsoft’s consoles. If PlayStation can have VR, then surely Xbox can too.

— Mat Smith

The biggest stories you might have missed

Grindr sued for allegedly sharing users’ HIV status and other info with ad companies

What we watched: Bluey’s joyful finales

Amazon halts drone deliveries in California, but kicks off tests in Phoenix

​​You can get these reports delivered daily direct to your inbox. Subscribe right here!

The EU opens an investigation into TikTok Lite, citing addiction concerns

This follows a more comprehensive probe from February.

The European Union (EU) has opened a second formal investigation into TikTok. The probe involves the addictive nature of TikTok Lite, a smaller version of the app that takes up less memory and was built to perform over slower internet connections. The wrinkle might be a design aspect that allows users to earn points by watching and liking videos. These points can be exchanged for TikTok’s proprietary digital currency and even Amazon vouchers. The EU’s Commission has expressed concern that this type of “task and reward” design language could impact the mental health of young users by “stimulating addictive behavior.”

Continue reading.

Russian court sentences Meta spokesperson in absentia to six years in prison

Andy Stone was convicted after ‘publicly defending terrorism.’

A few months after Russian officials placed Meta spokesperson Andy Stone on a wanted list and started a criminal investigation, a Moscow court has issued an arrest warrant for him on several terrorism-related charges in February. It cited Stone’s alleged “promotion of terrorist activities, public calls for terrorist activities, public justification of terrorism or propaganda of terrorism and public calls for extremist activities.”

Russia’s investigative committee opened a probe into Meta in March 2022. It claimed Stone had incited extremist activity after lifting “a ban on calls for violence against the Russian military on its platforms.” Stone said Meta was “temporarily” allowing some posts to stay on its platforms that would have previously been taken down for inciting violence, but noted the company would still outlaw “credible calls for violence against Russian civilians.”

Continue reading.

Embracer splits up its messy gaming empire into three companies

They’ll specialize in AAA, tabletop and indie titles.

Embracer

The company has announced plans to split into three separate parts. The first is Middle-earth Enterprises & Friends, specializing in AAA games like Tomb Raider and Dead Island – and, of course, anything Lord of the Rings. Asmodee will handle the tabletop gaming segment, which includes Ticket to Ride, 7 Wonders, Azul, CATAN, Dobble and Exploding Kittens. Coffee Stain & Friends will be the company’s indie-centric group, with properties including Deep Rock Galactic and Goat Simulator.

Continue reading.

Atlus first teased that it was working on a new RPG with a fantasy setting in mid-2023 — it also said way back then that it will be available sometime this year. Now, the developer has revealed that the game, Metaphor: ReFantazio, will come out on October 11 at a special livestream event. Katsura Hashino, the director of the game, as well as of Persona 3, 4 and 5, also introduced a 30-minute hands-on gameplay that gives you quite a lengthy look at its story and combat mechanics. Similar to the Persona games, Metaphor: ReFantazio has a turn-based combat system with what Atlus says is a "blend of real-time action."

Also, like the Persona games, you'll have to manage your time, so that you can build bonds with your allies and increase your "virtues" outside of dungeon-crawling. Metaphor: ReFantazio is set in the fictional United Kingdom of Euchronia, which was plunged into chaos after the assassination of its king. In the middle of the royal tournament for the throne, the protagonist and his partner fairy Galica go on a journey to find the cursed prince that was thought to be dead and team up with new allies along the way. 

Physical copies of the game, both standard and limited Collector's editions, are now available for pre-order, but you'll have to wait a bit if you'd rather get the digital version. Upon launch, the game will be available for various consoles, namely the Xbox Series X|S, PlayStation 5 and PlayStation 4, as well as on Windows and Steam on PC.

If the 30-minute gameplay footage is too long for you, here's a new trailer you can watch instead:

Microsoft has unveiled its latest light AI model called the Phi-3 Mini designed to run on smartphones and other local devices, it revealed in a new research paper. Trained on 3.8 billion parameters, it's the first of three small Phi-3 language models the company will release in the near future. The aim is to provide a cheaper alternative to cloud-powered LLMs, allowing smaller organizations to adopt AI. 

According to Microsoft, the new model handily outperforms its previous Phi-2 small model and is on par with larger models like Llama 2. In fact, the company says Phi-3 Mini provides responses close to the level of a model 10 times its size. 

"The innovation lies entirely in our dataset for training," according to the research paper. That dataset is based on the Phi-2 model, but uses "heavily filtered web data and synthetic data," the team states. In fact, a separate LLM was used to do both of those chores, effectively creating new data that allows the smaller language model to be more efficient. The team was supposedly inspired by children's books that use simpler language to get across complex topics, according to The Verge. 

Microsoft

While it still can't produce the results of cloud-powered LLMs, Phi-3 Mini can outperform Phi-2 and other small language models (Mistral, Gemma, Llama-3-In) in tasks ranging from math to programming to academic tests. At the same time, it runs on devices as simple as smartphones, with no internet connection required.

Its main limitation is breadth of "factual knowledge" due to the smaller dataset size — hence why it doesn't perform well in the "TriviaQA" test. Still, it should be good for models like that only require smallish internal data sets. That could allow companies that can't afford cloud-connected LLMs to jump into AI, Microsoft hopes.

Phi-3 Mini is now available on Azure, Hugging Face and Ollama. Microsoft is next set to release Phi-3 Small and Phi-3 Medium with significantly higher capabilities (7 billion and 14 billion parameters, respectively). 

Nearly a year after adding generative AI-powered editing capabilities to Photoshop, Adobe is souping up its flagship product with even more AI. On Tuesday, the company announced that Photoshop is getting the ability to generate images with simple text prompts directly within the app. There are also new features to let the AI draw inspiration from reference images to create new ones and generate backgrounds more easily. The tools will make using Photoshop easier for both professionals as well as casual enthusiasts who may have found the app’s learning curve to be steep, Adobe thinks.

“A big, blank canvas can sometimes be the biggest barrier,” Erin Boyce, Photoshop’s senior marketing director, told Engadget in an interview. “This really speeds up time to creation. The idea of getting something from your mind to the canvas has never been easier.” The new feature is simply called “Generate Image” and will be available as an option in Photoshop right alongside the traditional option that lets you import images into the app.

An existing AI-powered feature called Generative Fill that previously let you add, extend or remove specific parts of an image has been upgraded too. It now allows users to add AI-generated images to an existing image that blend in seamlessly with the original. In a demo shown to Engadget, an Adobe executive was able to circle a picture of an empty salad dish, for instance, and ask Photoshop to fill it with a picture of AI-generated tomatoes. She was also able to generate variations of the tomatoes and choose one of them to be part of the final image. In another example, the executive replaced an acoustic guitar held by an AI-generated bear with multiple versions of electric guitars just by using text prompts, and without resorting to Photoshop’s complex tools or brushes.

Adobe

These updates are powered by Firefly Image 3, the latest version of Adobe’s family of generative AI models that the company also unveiled today. Adobe said Firefly 3 produces images of a higher quality than previous models, provides more variations, and understands your prompts better. The company claims that more than 7 billion images have been generated so far using Firefly.

Adobe is far from the only company stuffing generative AI features into its products. Over the last year, companies, big and small, have revamped up their products and services with AI. Both Google and Microsoft, for instance, have upgraded their cash cows, Search and Office respectively, with AI features. More recently, Meta has started putting its own AI chatbot into Facebook, Messenger, WhatsApp, and Instagram. But while it’s still unclear how these bets will pan out, Adobe’s updates to Photoshop seem more materially useful for creators. The company said Photoshop’s new AI features had driven a 30 percent increase in Photoshop subscriptions.

Meanwhile, generative AI has been in the crosshairs of artists, authors, and other creative professionals, who say that the foundational models that power the tech were trained on copyrighted media without consent or compensation. Generative AI companies are currently battling lawsuits from dozens of artists and authors. Adobe says that Firefly was trained on licensed media from Adobe Stock, since it was designed to create content for commercial use, unlike competitors like Midjourney whose models are trained in part by illegally scraping images off the internet. But a recent report from Bloomberg showed that Firefly, too, was trained, in part, on AI-generated images from the same rivals including Midjourney (an Adobe spokesperson told Bloomberg that less than 5 percent of images in its training data came from other AI rivals).

To address concerns about the use of generative AI to create disinformation, Adobe said that all images created in Photoshop using generative AI tools will automatically include tamper-proof “Content Credentials”, which act like digital “nutrition labels” indicating that an image was generated with AI, in the file’s metadata. However, it's still not a perfect defense against image misuse, with several ways to sidestep metadata and watermarks. 

The new features will be available in beta in Photoshop starting today and will roll out to everyone later this year. Meanwhile, you can play with Firefly 3 on Adobe’s website for free. 

Amazon customers in California won't be able to get drone deliveries anymore. The e-commerce company has closed its delivery site in Lockeford, which has been operational since 2022, and will now offer its personnel in the area opportunities at other sites. Amazon made the revelation almost as an aside in an announcement that it's launching drone deliveries in the West Valley Phoenix Metro area later this year. Its drones will be deployed from facilities near its Tolleson fulfillment center. Amazon says it's the first time drone deliveries will be fully integrated into its network, and it will allow the company to fulfill and deliver purchases more quickly. 

The company doesn't have an exact launch date for its drone deliveries in Phoenix, because it's still working with the Federal Aviation Administration (FAA) and local officials to get the permits it needs. It does have the support of Phoenix Mayor Kate Gallego, though, who called drone deliveries "the future" and said it would help her city "reduce local pollution" and further cement it "as a hotbed for the innovative technology of tomorrow."

While Amazon's drone delivery operations are shutting down in California, it'll continue its activities in College Station, Texas. Shortly after it started using drones as couriers in those two areas, The Information reported that the company has made just a handful of deliveries via the method, mostly due to FAA limitations that prohibit the machines from flying over roads or people unless Amazon gets permission for every case. It eventually reached 100 drone deliveries by the middle of 2023, though that was likely far from what the company had hoped to get by then, since it aimed to reach 10,000 deliveries by the end of the year. 

Those setbacks, however, don't seem to have deterred Amazon. It's currently testing its next-gen MK30 drones that can fly twice as far as its current drones, and it also said that it's deploying drone deliveries in more locations in the US next year. 

Newsletter platform Ghost is the latest service to pledge support for ActivityPub, the open source protocol powering the fediverse. The company announced Monday it would add ActivityPub support later this year in a move that could bring tens of millions of people into the fediverse.

The fediverse is a growing collection of services, including Mastodon, Flipboard and Threads, that support the ActivityPub protocol. It’s part of a growing movement for decentralized social media services, which rely on open protocols rather than closed networks. Proponents often compare it to email, which allows people to communicate regardless of their preferred app or platform.

In a blog post laying out its vision, Ghost said it was joining the fediverse in an effort to “bring back” the open web. “On, Ghost publishers will be able to follow, like and interact with one another in the same way that you would normally do on a social network — but on your own website,” the company wrote. “The difference, of course, is that you’ll also be able to follow, like, and interact with users on Mastodon, Threads, Flipboard, Buttondown, WriteFreely, Tumblr, WordPress, PeerTube, Pixelfed... or any other platform that has adopted ActivityPub, too.”

While Ghost says ActivityPub integration will be optional for publishers, the company notes that its entry into the fediverse could bring "tens of millions" of new people into the space. A number of popular newsletters run on Ghost, including Platformer, Garbage Day, She’s a Beast, as does the independent tech news site 404 Media.

A Russian military court sentenced Meta spokesperson Andy Stone in absentia to six years in prison for "publicly defending terrorism," Reuters reports. Stone's lawyer reportedly asked for an acquittal and there are plans to appeal the sentence. 

A few months after Russian officials placed him on a wanted list and started a criminal investigation, a Moscow court issued an arrest warrant for Stone on several terrorism-related charges in February. It cited Stone's alleged "promotion of terrorist activities, public calls for terrorist activities, public justification of terrorism or propaganda of terrorism and public calls for extremist activities."

The measure follows Russia's investigative committee opening a probe into Meta in March 2022. It claimed that Stone had incited extremist activity after lifting "a ban on calls for violence against the Russian military on its platforms." Around that time, Stone said Meta was "temporarily" allowing some posts that would have previously been taken down for inciting violence to stay on its platforms, but noted that the company would still outlaw “credible calls for violence against Russian civilians.”

In any case, it seems unlikely that Stone will actually spend time behind bars in Russia, unless he were to travel there or to a country that has an extradition treaty with the nation. It's not uncommon for a person to be charged or sentenced (often for spying- or hacking-related crimes) in another country and never actually have to deal with those consequences.

Russia has designated Meta as an extremist organization. It blocked access to Facebook and Instagram soon after commencing its invasion of Ukraine in 2022.

Engadget has contacted Meta for comment.

In 2024, four billion people — about half the world’s population — in 64 countries including large democracies like the US and India, will head to the polls. Social media companies like Meta, YouTube and TikTok, have promised to protect the integrity of those elections, at least as far as discourse and factual claims being made on their platforms are concerned. Missing from the conversation, however, is closed messaging app WhatsApp, which now rivals public social media platforms in both scope and reach. That absence has researchers from non-profit Mozilla worried.

“Almost 90% of the safety interventions pledged by Meta ahead of these elections are focused on Facebook and Instagram,” Odanga Madung, a senior researcher at Mozilla focused on elections and platform integrity, told Engadget. “Why has Meta not publicly committed to a public road map of exactly how it’s going to protect elections within [WhatsApp]?”

Over the last ten years, WhatsApp, which Meta (then Facebook) bought for $19 billion in 2014, has become the default way for most of the world outside the US to communicate. In 2020, WhatsApp announced that it had more than two billion users around the world — a scale that dwarfs every other social or messaging app except Facebook itself.

Despite that scale, Meta’s focus has mostly been only on Facebook when it comes to election-related safety measures. Mozilla’s analysis found that while Facebook had made 95 policy announcements related to elections since 2016, the year the social network came under scrutiny for helping spread fake news and foster extreme political sentiments. WhatsApp only made 14. By comparison, Google and YouTube made 35 and 27 announcements each, while X and TikTok had 34 and 21 announcements respectively. “From what we can tell from its public announcements, Meta’s election efforts seem to overwhelmingly prioritize Facebook,” wrote Madung in the report.

Mozilla is now calling on Meta to make major changes to how WhatsApp functions during polling days and in the months before and after a country’s elections. They include adding disinformation labels to viral content (“Highly forwarded: please verify” instead of the current “forwarded many times), restricting broadcast and Communities features that let people blast messages to hundreds of people at the same time and nudging people to “pause and reflect” before they forward anything. More than 16,000 people have signed Mozilla’s pledge asking WhatsApp to slow the spread of political disinformation, a company spokesperson told Engadget.

WhatsApp first started adding friction to its service after dozens of people were killed in India, the company’s largest market, in a series of lynchings sparked by misinformation that went viral on the platform. This included limiting the number of people and groups that users could forward a piece of content to, and distinguishing forwarded messages with “forwarded” labels. Adding a “forwarded” label was a measure to curb misinformation — the idea was that people might treat forwarded content with greater skepticism.

“Someone in Kenya or Nigeria or India using WhatsApp for the first time is not going to think about the meaning of the ‘forwarded’ label in the context of misinformation,” Madung said. “In fact, it might have the opposite effect — that something has been highly forwarded, so it must be credible. For many communities, social proof is an important factor in establishing the credibility of something.”

The idea of asking people to pause and reflect came from a feature that Twitter once implemented where the app prompted people to actually read an article before retweeting it if they hadn’t opened it first. Twitter said that the prompt led to a 40% increase in people opening articles before retweeting them

And asking WhatsApp to temporarily disable its broadcast and Communities features arose from concerns over their potential to blast messages, forwarded or otherwise, to thousands of people at once. “They’re trying to turn this into the next big social media platform,” Madung said. “But without the consideration for the rollout of safety features.”

“WhatsApp is one of the only technology companies to intentionally constrain sharing by introducing forwarding limits and labeling messages that have been forwarded many times,” a WhatsApp spokesperson told Engadget. “We’ve built new tools to empower users to seek accurate information while protecting them from unwanted contact, which we detail on our website.”

Mozilla’s demands came out of research around platforms and elections that the company did in Brazil, India and Liberia. The former are two of WhatsApp’s largest markets, while most of the population of Liberia lives in rural areas with low internet penetration, making traditional online fact-checking nearly impossible. Across all three countries, Mozilla found political parties using WhatsApp’s broadcast feature heavily to “micro-target” voters with propaganda, and, in some cases, hate speech.

WhatsApp’s encrypted nature also makes it impossible for researchers to monitor what is circulating within the platform’s ecosystem — a limitation that isn’t stopping some of them from trying. In 2022, two Rutgers professors, Kiran Garimella and Simon Chandrachud visited the offices of political parties in India and managed to convince officials to add them to 500 WhatsApp groups that they ran. The data that they gathered formed the basis of an award-winning paper they wrote called “What circulates on Partisan WhatsApp in India?” Although the findings were surprising — Garimella and Chandrachud found that misinformation and hate speech did not, in fact, make up a majority of the content of these groups — the authors clarified that their sample size was small, and they may have deliberately been excluded from groups where hate speech and political misinformation flowed freely.

“Encryption is a red herring to prevent accountability on the platform,” Madung said. “In an electoral context, the problems are not necessarily with the content purely. It’s about the fact that a small group of people can end up significantly influencing groups of people with ease. These apps have removed the friction of the transmission of information through society.”