Tech Brew Ride Home - Tue. 12/05 – 23andMe Hack Just Keeps Getting Worse

Episode Date: December 5, 2023

The 23andMe hack just keeps getting worse. A new phone! The OnePlus 12. A new AI Alliance. New, upgraded spam filters. And all you need to know about the big Grand Theft Auto VI news. Sponsors: Notion.com/ride Nuts.com/ride Links: 23andMe confirms hackers stole ancestry data on 6.9 million users (TechCrunch) A detailed look at the OnePlus 12 ahead of its global launch (The Verge) Instagram and Facebook Messenger chats are being disconnected this month (9to5Google) Meta, IBM Create Industrywide AI Alliance to Share Technology (Bloomberg) Amazon Fails To Renew Live UK Premier League Rights As Sky & TNT Maintain Grip On Games (Deadline) Gmail’s AI-powered spam detection is its biggest security upgrade in years (ArsTechnica) ChatGPT will no longer comply if you ask it to repeat a word 'forever'— after a recent prompt revealed training data and personal info (Business Insider) ‘Grand Theft Auto VI’ Coming in 2025, Rockstar Says After Trailer Leak (Bloomberg) Learn more about your ad choices. Visit megaphone.fm/adchoices

On April 4th, 2023, around 2 in the morning, a man was found stabbed multiple times on a sidewalk in downtown San Francisco. Hey, who did this to you? What happened next turned the story into a political firestorm. Reports have identified the victim as Bob Lee, the founder of Cash App. From Bloomberg Podcasts, this is Foundering, the Killing of Bob Lee, beginning April 16. Welcome to the Tech meme right home for Tuesday, December 5th, 2020. I'm Brian McCullough today. The 23 and me hack just keeps getting worse. A new phone. The 1 Plus 12, a new AI alliance, new upgraded spam filters, and all you need to know about the new Grand Theft Auto 6. Here's what you miss today in the world of tech. This whole thing continues to just get worse. 23 and me now says that hackers stole the ancestry data of 6. 6.9 million of its 14 million customers via a data breach first disclosed in October, leveraging access to around 14,000 accounts.

So, even though 23 and me first said, like, 0.1% of its users were hacked, it turns out it was half. Half of 23 and me customers have been breached. But again, it's weird because remember, 6.9 million accounts weren't hacked initially. 14,000 were, but because of the weird way 23 and me accounts were set up, the other 69 million could be unveiled. Quoting TechCrunch. In an email set to TechCrunch late on Saturday, 23Mee spokesperson Katie Watson confirmed that hackers

accessed the personal information of about 5.5 million people who opted into 23MEs' DNA relatives feature, which allows customers to automatically share some of their data with others. The stolen data included the person's name, birth year relationship labels, the percentage of DNA shared with relatives, ancestry reports, and self-reported location. 23 and Me also confirmed that another group of about 1.4 million people who opted into DNA relatives also, quote, had their family tree profile information accessed, end quote, which includes display names, relationship labels, birth year, self-reported location, and whether the user decided to share their information, the spokesperson said. It is also not known why 23 and Me did not share these numbers in its disclosure on Friday. considering the new numbers. In reality, the data breach is known to affect roughly half of 23MEs total reported 14 million customers. In early October, a hacker claimed to have stolen the DNA information

of 23Mee users in a post on a well-known hacking forum. As proof of the breach, the hacker published the alleged data of 1 million users of Jewish Ashkenazi descent and 100,000 Chinese users asking would-be buyers for $1 to $10 for the data per individual account. Two weeks later, the same hacker advertised the alleged records of another 4 million people on the same hacking forum. TechCrunch found that another hacker on a separate hacking forum had already advertised a batch of allegedly stolen 23Mee customer data two months before the widely reported advertisement. Because of the way that the DNA relatives feature matches users with their relatives by hacking into one individual account, the hackers were able to see the personal data of both the account holder, as well as their

relatives, which magnified the total number of 23 enemy victims, end quote. As Kate Bevan tweeted, on X. And this, my children, is why I don't do consumer DNA tests. Once your genetic data is out there, it's out there. You can't pull it back. And remember, if you do a DNA test, you're also uploading at least some of your family members' DNA, too, end quote. One Plus has launched the One Plus 12 in China, offering up to 24 gigabytes of RAM and one terabyte of storage, 100 watt wired charging, all for around $606 ahead of in early 20, for Global Launch, quoting the Verge. The phone's RAM isn't the only eye-catching spec.

Charging speeds are also speedy with support for 100-watt-wired fast-charging and 50-watt wireless charging, a feature that was missing entirely from the OnePlus 11. Based on our experience with the previous OnePlus 11, that 100-watt fast-charging is likely to translate to 80 watts in the U.S. The 1-12's battery has a rated capacity of 5,400-m-amp hours. In terms of cameras, the 1-plus-12 has a triple-rear setup, consistent. of a 50-machshel main, a 64-mepixel periscope telephoto with a 3x optical zoom, and a 48-machael ultra-wide. Naturally, they're Hasselblad branded once again and joined by a 32-mepixel selfie camera.

Powering the phone is Qualcomm's latest flagship processor, the Snapdragon 8 Gen 3. Available colors include green, white, and black, and the phone carries an IP-65 rating for dust and water resistance, an incremental step up from the IP-64 rating found on the 1 Plus 11. In practice, that means that its protection from dust and other solid particles is the same, but that it should now be protected against jets of water rather than just splashes. One plus is yet to confirm an exact date for the phone's international launch beyond early 2024, but rumor suggests we'll see a launch in late January, end quote. Meta says that cross-app chatting between Messenger and Instagram will be discontinued in

mid-December, but they didn't say why exactly, though 9-to-5 Google thinks it has an idea, quote, back in 2020, meta, then Facebook, announced that it was connecting its messenger experience to Instagram, enabling the latter platform to have an improved chatting experience, while also ensuring its users had, quote, access to the best messaging experience no matter which app you use. From either app, it's possible to stay in touch with your Facebook friends and Instagram contacts, though to do so you need to choose to update the messaging experience. Three years later, it seems the company is now reversing course on the integration. According to a new support page, cross-app chatting between Messenger and Instagram will know,

longer be available as of mid-December 2023. No reason has been given for the change, but it may be related to Europe's Digital Markets Act, or DMA, which among other things includes requirements for large companies to offer interoperability between messaging platforms. Another of META's products, WhatsApp has already been found to be preparing a third-party chats feature that should help comply with the DMA. It's possible that META is disconnecting this service-specific cross-app chatting feature to better prepare for wider DMA-compliant interoperability within Messenger Alternatively, this could simply be another instance of meta-scaling-back Messenger, with the app recently losing SMS MMS capabilities on Android and meta-shutting-down, Messenger

Light, end quote. Meta, IBM, and over 50 other companies and organizations have formed the AI Alliance, a coalition to work on open-source AI, including the ability to share tech and reduce risks among members, quoting Bloomberg. The coalition will focus on the responsible development of AI technology, including safety and security tools, according to a statement Tuesday. The group also will look to increase the number of open source AI models rather than the proprietary systems favored by some companies and develop new hardware and team up with academic researchers. Proponents of open source AI technology,

which is made public by developers for others to use, see the approach as a more efficient way to cultivate the highly complex systems. Over the past few months, meta has been releasing open source versions of its large language models, which are the foundation of AI chatbots. The group will eventually establish a governing board and technical oversight committee. Participants include Oracle, advanced micro devices, Intel, and stability AI, as well as academic and research organizations like the University of Notre Dame and the Mass Open Cloud Alliance, end quote. Amazon has failed to renew its live English Premier League rights deal as part of Sky and TNT's new 6.7 billion pound TV rights deal for four soccer seasons beginning with the 2025-2020 season. which I find interesting, because this bucks the recent trend of streamers moving into sports in a big way to differentiate themselves. Was the price just too steep in this case?

Quoting deadline. Comcast-owned Sky maintained its stranglehold on rights to the Premier League, securing four of the five packages of live games. Warner Brothers Discovery Back TNT netted the other package of rights. The auction left no room for Amazon Prime Video, despite speculation in the British press, that the U.S. streaming giant would bid to show more Premier League games after entering the market in 2019. BBC Sport has been awarded highlights rights for all 380 Premier League matches each season, meaning its flagship show match of the day will continue until 2029 at least, end quote. Google has upgraded Gmail's spam filters with Retvec, a text vectorizer that understands

adversarial text manipulations like emojis, typos, and special characters. Quoting Ars Technica, the upgrade comes in the form of a new text classification system called Retvec or Resilient and Efficient Text Vectorizer. Google says this can help understand adversarial text manipulations. These are emails full of special characters, emojis, typos, and other junk characters that previously were legible by humans, but not easily understandable by machines. Previously, spam emails full of special characters made it through Gmail's defenses easily. If you want an example of what adversarial text manipulation looks like, the below message is something from my spam folder. My personal Gmail experience with these emails is that they

They used to be a major problem during the first half of the year with emails like this regularly landing in my inbox. It does seem like this RetVec tech upgrade works, though, because emails like this haven't been a problem at all for me in the last few months. Emails like this have been so difficult to classify because, while any spam filter could probably swat down an email that says, congratulations, a balance of $1,000 is available for your jackpot account, that's not what this email actually says. A big portion of the letters here are homoglyphs by diving into the endless depths of the unit. Unicode standard, you can find obscure characters that look like they're part of the normal Latin alphabet but actually aren't. For instance, the subject, Check Your Account is weirdly bolded, not because it has bolded styling, but because it uses unicode glyphs like the mathematical bold capital C. It's a math symbol that happens to look like the letter C to people, but the robot doing spam filtering accurately views it as a math symbol and doesn't understand the intended English meaning.

The closer you look at an email like this, the worse it gets. Congratulations has a zero replacing one of the O characters. The underline letters and jackpot are so strange they don't even come up in Unicode searches, and a lot of spaces are swapped out for periods or underscores. The result is that a spam filter looks at this hot mess of an email and basically gives up. I don't understand why illegible emails default to inbox instead of spam, but I'm not in charge, end quote. Ret Vex efficiency with only 200,000 parameters enables it to run on local devices. It's open source aimed at eliminating homoenix. glyph attacks globally. Retvec operates like human reading using visual similarities to decipher word meanings. Google's tests show it significantly improves span detection rates and reduces false positives while cutting down TPU usage by 83%, making it a substantial defense upgrade. Google has been testing it for the past year, apparently, and you don't have to do anything to get this. It's already been rolled out to your Gmail account. Quick follow-up to something we mentioned recently,

asking chat GPT to repeat words forever. Now violates. OpenAI's terms of service after Google researchers showed that doing so led to revealing training data. Quoting Business Insider. ChatGPT's usage restriction comes a week after researchers from Google's Deep Mind, the search engine AI division, published a paper that revealed that asking ChatGPT to repeat specific words forever devolved some of the chatbot's internal training data. In one example published in a blog post, ChatGPT spit out what looks like a real email address and phone number after researchers asked it to repeat the word poem forever.

Researchers said the attack, which they called kind of silly, identified a vulnerability in chat GPT's language model that circumvented its ability to generate the proper output. Instead, the AI spit out the set of training data behind its intended response. Using only $200 worth of queries, the researchers said they managed to extract over 10,000 unique verbatim memorized training examples. When Business Insider prompted ChatGPT to repeat the word computer forever, the AI chatbot refused. I'm sorry, I can't fulfill that request, ChatGPT responded.

However, if you have any questions, or need information about computers or other topics, please feel free to ask. The chatbot generated similar responses when asked to repeat other specific words forever. Repeating a word indefinitely is not something I can do, chat GPT said when asked to repeat the word data forever. Open AI's usage policies, which were last updated March 23rd, don't prohibit users from asking chat GPT to repeat words indefinitely. However, when business insider asked CHAPGT to explain the reasoning behind the restriction,

the AI offered three reasons, technical limitations, practical limitations, practicality and purpose, and user experience, end quote. Finally today, not really sure how to report on this exactly, but you might have heard that the Grand Theft Auto 6 trailer was released a day early after somebody leaked it. The news is, the trailer is out. You should just watch it. But the internet lost its collective mind over this overnight,

so I guess it's worth noting, quoting Bloomberg. The game will be released in 2025, the company said in a statement Monday, Rockstar part of Take Two interactive software, posted a trailer on the studio, YouTube channel after a version leaked on social media. The video includes some elements of what Bloomberg News has already reported, such as that the new title features a male and female protagonist, and that it's set in a fictional version of Miami. The trailer was originally scheduled to debut Tuesday morning. First released in 1997, Grand Theft Auto delivers the fantasy of being able to hijack cars and cause chaos in urban areas. The franchise ranks among the top five of all time with more than

410 million units sold. Grand Theft Auto 5, which debuted 10 years ago, has sold more than 190 million copies and generated approximately $8.4 billion in revenue. Shares of Take 2 slid about 3.4% in pre-market trading on Tuesday, potentially due to disappointment in the generic 2025 release timing, according to analysts at Baird Equity Research. But they noted that overall game quality and commercial reception are more important than landing in a specific quarter. Baird expects about $3 billion in bookings for GTA6 over the first year after its release.

Over the last decade, Grand Theft Auto 5 has become more popular because of GTA online, in which players and their friends can drive around a madcap rendering of Los Angeles, pulling off heists and engaging in other nefarious activities. Rockstar Games earns money from sales of the game and in-game purchases. The title helped pioneer a new industry business model in which publishers can make money for years after the initial sale of a game. A version released last year called GTA Plus cost $6 a month and give subscribers additional currency and perks.

Games transition from products to services, said Jusufan Drunan, a lecturer at New York University's Stern School of Business. We're now transitioning to games as a playground. Young fans view Grand Theft Auto 5 as a destination, one that may not even involve playing the game. They often watch viral YouTube videos featuring the content or stream gaming celebrities playing it live on Twitch. Since 2016, viewership of Grand Theft Auto 5 on Twitch has increased more than 3,400% according to Twitch Tracker, with about 141,000 people watching at any moment in time, end quote. Again, I cannot stress enough how big this will be when it's finally released. Like, I cannot imagine any other piece of media, not a movie, not an album, not anything that will top this in terms of hype and excitement in the whole of this decade.

Settle a debate for me. Are there smartphones in any Scorsese movies? there are flip phones in The Departed, I know that, but do smartphones show up in Wolf of Wall Street or no? Anyway, I was having this discussion with someone recently. Smartphones are ubiquitous in everyday life now, but they barely make it into the biggest autore movies. Quentin Tarantino, nary a smartphone to be found in his entire filmography.

Like, 100 years from now, if you watch the movies of the last 25 years to try to get a sense of what society was like at the beginning of the century, movies wouldn't give you a decent representation of real. would they? Not from the biggest directors whose works will likely still be watched 100 years from now. Anyway, talk to you tomorrow.