23andMe told victims of data breach that suing is futile, letter shows

Victims are still arguing that 23andMe’s security measures were inadequate.

23andMe told victims of data breach that suing is futile, letter shows

Enlarge (credit: Bloomberg / Contributor | Bloomberg)

23andMe is "shamelessly" blaming victims of a data breach impacting 6.9 million users, a lawyer representing victims pursuing a class-action lawsuit, Hassan Zavareei, told TechCrunch.

Zavareei shared a letter from 23andMe lawyers that urged users suing to "consider the futility of continuing to pursue an action in this case," because their claims are allegedly meritless and "the information that was potentially accessed cannot be used for any harm."

Last year, hackers accessed 14,000 accounts on 23andMe by using passwords that had been previously breached during security incidents on other websites. By using this tactic, known as credential stuffing, hackers could access the personal data of millions of 23andMe users who opted into a DNA Relatives feature, including genetic information like the percentage of DNA shared with compromised users.

Read 24 remaining paragraphs | Comments

All Science journals will now do an AI-powered check for image fraud

It will only catch the most blatant problems, but it’s definitely overdue.

All Science journals will now do an AI-powered check for image fraud

(credit: Aurich Lawson)

On Thursday, the research publisher Science announced that all of its journals will begin using commercial software that automates the process of detecting improperly manipulated images. The move comes many years into our awareness that the transition to digital data and publishing has made it comically easy to commit research fraud by altering images.

While the move is a significant first step, it's important to recognize the software's limitations. While it will catch some of the most egregious cases of image manipulation, enterprising fraudsters can easily avoid being caught if they know how the software operates. Which, unfortunately, we feel compelled to describe (and, to be fair, the company that has developed the software does so on its website).

Fantastic fraud and how to catch it

Much of the image-based fraud we've seen arises from a dilemma faced by many scientists: It's not a problem to run experiments, but the data they generate often isn't the data you want. Maybe only the controls work, or maybe the experiments produce data that is indistinguishable from controls. For the unethical, this doesn't pose a problem since nobody other than you knows what images come from which samples. It's relatively simple to present images of real data as something they're not.

Read 15 remaining paragraphs | Comments

Qualcomm’s XR2+ Gen 2 SoC sets up a wave of Apple Vision Pro competitors

Qualcomm says five companies are working on new headsets.

Qualcomm's XR reference device.

Enlarge / Qualcomm's XR reference device. (credit: Qualcomm)

The Apple Vision Pro is coming out sometime in early 2024, and since it is a VR/AR headset that runs iOS apps, Team Android would like to have a competitor available. Samsung, Google, and Qualcomm vaguely announced a mixed-reality partnership a year ago, which would have Qualcomm building chips, Google building software, and Samsung shipping products. Step 1 of this partnership has been announced: The Qualcomm Snapdragon XR2+ SoC, which will power many of these headsets from Samsung and others.

Perhaps more interesting than the individual specs of Qualcomm's chip is all the partner talk surrounding the launch. Samsung says it's "thrilled to collaborate with Qualcomm Technologies and Google in revolutionizing the mobile industry once more" and promises a "best-in-class XR experience for Galaxy users." Google is promising an "immersive and spatial XR" experience based on Android. Qualcomm is making a reference headset design based on the chip and says at least five companies have hardware in development, including Samsung, Google, HTC, Immersed, and Play for Dream.

The XR2+ Gen 2 is an upgrade over the XR2 Gen 2 (non-plus) that came out in 2023 and shipped in the Meta Quest 3. The most significant difference is a bump in resolution, with the plus model supporting "4.3K" resolution per eye at 90 fps. The old chip only did 3K by 3K. The resolution of these headsets is critical, not just because games will look nicer, but if you're trying to display a virtual desktop and read text, you're pushing the resolution limits of current headsets.

Read 2 remaining paragraphs | Comments

Banana Pi BPI-M6 single-board PC with quad-core ARM Cortex-A73 CPU and NPU for on-device AI features

The Banana Pi BPI-M6 is a credit card-sized computer that looks like a Raspberry Pi at first glance. But this little computer has a few special tricks up its sleeve. Among other things, it has an integrated NPU with up to 6.75 TOPS of performance for …

The Banana Pi BPI-M6 is a credit card-sized computer that looks like a Raspberry Pi at first glance. But this little computer has a few special tricks up its sleeve. Among other things, it has an integrated NPU with up to 6.75 TOPS of performance for AI tasks, two micro HDMI ports (one for output and […]

The post Banana Pi BPI-M6 single-board PC with quad-core ARM Cortex-A73 CPU and NPU for on-device AI features appeared first on Liliputing.

Symmetrische Datenraten von 10 GBit/s: Glasfaserbetreiber setzt bereits komplett auf XGS-PON

10 Gigabit durch XGS-PON bieten noch wenige Glasfaserzugänge. Vattenfall Eurofiber ist eher Experten bekannt und setzt in seinem Glasfasernetz von Anfang an auf den Standard. (Vattenfall, Glasfaser)

10 Gigabit durch XGS-PON bieten noch wenige Glasfaserzugänge. Vattenfall Eurofiber ist eher Experten bekannt und setzt in seinem Glasfasernetz von Anfang an auf den Standard. (Vattenfall, Glasfaser)

Mathematik & Neurologie: Unser Gehirn verarbeitet die Zahlen unterschiedlich

Es scheint bei Neuronen für niedrige Zahlen einen hemmenden Mechanismus zu geben, der andere Neuronen unterdrückt. Dadurch fallen Schätzungen bis vier meist richtig aus.  (Medizin, Wissenschaft)

Es scheint bei Neuronen für niedrige Zahlen einen hemmenden Mechanismus zu geben, der andere Neuronen unterdrückt. Dadurch fallen Schätzungen bis vier meist richtig aus.  (Medizin, Wissenschaft)