Connect with us

News

Scientists create AI neural net that can unlock digital fingerprint-secured devices

Fingerprint scan. | Credit: RCPA

Published

on

Computer scientists at New York University and Michigan State University have trained an artificial neural network to create fake digital fingerprints that can bypass locks on cell phones. The fakes are called “DeepMasterPrints”, and they present a significant security flaw for any device relying on this type of biometric data authentication. After exploiting the weaknesses inherent in the ergonomic needs of cellular devices, DeepMasterPrints were able to imitate over 70% of the fingerprints in a testing database.

An artificial neural network is a type of artificial intelligence comprising computer algorithms modeled after the human brain’s ability to recognize patterns. The DeepMasterPrints system was trained to analyze sets of fingerprint images and generate a new image based on the features that occurred most frequently. This “skeleton key” could then be used to exploit the way cell phones authenticate user fingerprints.

In cell phones, the necessarily small size of fingerprint readers creates a weakness in the way they verify a print. In general, phone sensors only capture a partial image of a print when a user is attempting to unlock the device, and that piece is then compared to the phone’s authorized print image database. Since a partial print means there are fewer characteristics to distinguish it than a full print, a DeepMasterPrint needs to match fewer features to imitate a fingerprint. It’s worth noting that the concept of exploiting this flaw is not unique to this particular study; however, generating unique images rather than using actual or synthesized images is a new development.

An overview of the DeepMasterPrint system. | Credit: IEEE

The team involved in the study resulting in the DeepMasterPrint creation initiated it as part of the ongoing assessment of vulnerabilities in fingerprint recognition systems. Finding exploitable flaws and fixing them is a constant battle in all digital systems with a security component. With this reality in mind, the scientists determined that merely exposing the flaws of fingerprint systems would not provide an effective solution; a working example of how attacks could be executed provides more specific data for researchers to design around and protect against. Creating the DeepMasterPrint system was meant to address this need.

The results revealed by the DeepMasterPrint system are concerning for anyone relying on fingerprint authentication on their smartphones. Scientists compared the generated fake prints against templates generated by VeriFinger 9.0 SDK, Bozorth3, and Innovatrics IDKit 5.3 SKD, all of which are software systems used in fingerprint authentication systems worldwide. At a low false match rate, i.e., strict match requirements for authentication, the fake print generated by DeepMasterPrint could imitate 23% of the fingerprints in the test database. At a slightly higher false match rate that was still within standard phone authentication limits, the fake print imitated 77% of the test fingerprints.

Real fingerprints on the left vs. DeepMasterPrint generated fingerprints on the right. | Credit: IEEE

The scientists in this study did not create physical fingerprints to try and unlock actual phones, leaving that work to be done in the near future. However, even though the successful DeepMasterPrints have yet to be tested in true applications rather than a virtual environment, the data gathered confirmed the initial security concerns which inspired the experiment. Fingerprints are being used as identity verification in a growing number of applications beyond cell phone security, i.e., unlocking entryways, payment authentication, etc. The DeepMasterPrint system is another tool to help researchers guard their security as biometric authentication continues to expand.

Advertisement

Accidental computer geek, fascinated by most history and the multiplanetary future on its way. Quite keen on the democratization of space. | It's pronounced day-sha, but I answer to almost any variation thereof.

Advertisement
Comments

News

Nvidia CEO Jensen Huang regrets not investing more in Elon Musk’s xAI

The CEO stated that Nvidia is already an investor in xAI, but he wished he had given the artificial intelligence startup more money.

Published

on

Credit: Elon Musk/X

Nvidia CEO Jensen Huang revealed that one of his investment regrets is not putting more money into Elon Musk’s artificial intelligence startup, xAI. 

Speaking in a CNBC interview, Huang said Nvidia is already an investor in xAI but wished he had given the artificial intelligence startup more money. This was due to Musk’s record of building transformative companies such as Tesla and SpaceX.

A new wave of transformative AI firms

Huang said he’s very excited about xAI’s latest financing round. He described Musk’s company as part of a powerful new generation of AI developers, alongside OpenAI and Anthropic. that are reshaping the computing landscape.

“I’m super excited about the financing opportunity they’re doing. The only regret I have about xAI, we’re an investor already, is that I didn’t give him more money. You know almost everything that Elon’s pat of, you really want to be part of as well,” the Nvidia CEO stated.

The CEO also clarified Nvidia’s investment in xAI, revealing that Elon Musk had offered the investment opportunity to the chipmaker. “He (Musk) gave us the opportunity to invest in xAI. I’m just delighted by that,” Huang stated.

Advertisement

AI investment boom

Huang contrasted today’s AI-driven economy with the early days of the internet. “Back then, all the internet companies combined were maybe $30 or $40 billion in size,” he said. “If you look at the hyperscalers now, that’s about $2.5 trillion of business already operating today.”

He also stated that the ongoing shift from CPU-based computing to GPU-powered generative AI represents a “multi-trillion-dollar buildout” that Nvidia is looking to support. Huang added that every Nvidia engineer now works with AI coding assistants such as Cursor, which he called his “favorite enterprise AI service,” and it has led to a major productivity boost across the company.

Watch Nvidia CEO Jensen Huang’s CNBC interview in the video below.

Continue Reading

Investor's Corner

Stifel raises Tesla price target by 9.8% over FSD, Robotaxi advancements

Stifel also maintained a “Buy” rating for the electric vehicle maker.

Published

on

Credit: Tesla China

Investment firm Stifel has raised its price target for Tesla (NASDAQ:TSLA) shares to $483 from $440 over increased confidence in the company’s self-driving and Robotaxi programs. The new price target suggests an 11.5% upside from Tesla’s closing price on Tuesday.

Stifel also maintained a “Buy” rating despite acknowledging that Tesla’s timeline for fully unsupervised driving may be ambitious.

Building confidence

In a note to clients, Stifel stated that it believes “Tesla is making progress with modest advancements in its Robotaxi network and FSD,” as noted in a report from Investing.com. The firm expects unsupervised FSD to become available for personal use in the U.S. by the end of 2025, with a wider ride-hailing rollout potentially covering half of the U.S. population by year-end.

Stifel also noted that Tesla’s Robotaxi fleet could expand from “tiny to gigantic” within a short time frame, possibly making a material financial impact to the company by late 2026. The firm views Tesla’s vision-based approach to autonomy as central to this long-term growth, suggesting that continued advancements could unlock new revenue streams across both consumer and mobility sectors.

Tesla’s FSD goals still ambitious

While Stifel’s tone remains optimistic, the firm’s analysts acknowledged that Tesla’s aggressive autonomy timeline may face execution challenges. The note described the 2025 unsupervised FSD target as “a stretch,” though still achievable in the medium term.

Advertisement

“We believe Tesla is making progress with modest advancements in its Robotaxi network and FSD. The company has high expectations for its camera-based approach including; 1) Unsupervised FSD to be available for personal use in the United States by year-end 2025, which appears to be a stretch but seems more likely in the medium term; 2) that it will ‘probably have ride hailing in probably half of the populations of the U.S. by the end of the year’,” the firm noted.

Continue Reading

News

Tesla Cybertruck gets Full Self-Driving v14 release date, sort of

Published

on

Tesla Cybertruck owners are wondering when they will get access to the company’s Full Self-Driving version 14.1 that rolled out to other owners today for the first time.

Cybertruck owners typically receive Full Self-Driving updates slightly later than other drivers, as the process for the all-electric pickup is different. It is a larger vehicle that requires some additional attention from Tesla before FSD versions are rolled out, so they will be slightly delayed. CEO Elon Musk said the all-wheel steering technically requires a bit more attention before rollout as well.

After some owners got access to the v14.1 Full Self-Driving suite this morning, Cybertruck owners sought out a potential timeframe for when they would be able to experience things for themselves.

Tesla owners show off improvements with new Full Self-Driving v14 rollout

They were able to get an answer from Ashok Elluswamy, Tesla’s Head of AI, who said:

“We got you. Coming soon.”

The release of FSD v14.1 for Cybertruck will not be tempered, either. Elluswamy then confirmed that Tesla would be rolling out the full-featured FSD v14 for the pickup, meaning it would be able to reverse and park itself, among other features.

Elluswamy said it would be capable of these features, which were void in other FSD releases for Cybertruck in the past.

Tesla’s rollout of FSD v14.1 brings several extremely notable changes and improvements to the suite, including more refined operation in parking garages, a new ability to choose parking preferences upon arriving at your destination, a new driving mode called “Sloth,” which is even more reserved than “Chill,” and general operational improvements.

Those who were lucky enough to receive the suite have already started showing off the improvements, and they definitely seem to be a step up from what v13’s more recent versions were capable of.

CEO Elon Musk called v14 “sentient” a few weeks back, and it seems that it is moving toward that. However, he did state that additional releases with more capabilities would be available in the coming weeks, but many owners are still waiting for this first version.

Continue Reading

Trending