philip lelyveld The world of entertainment technology

Let’s not make the same mistakes with AI that we made with social media

PhilNote: this story is notable for its authors; Harvard's Bruce Schneier and Nathan Sanders

...

There is a lot we can learn about social media’s unregulated evolution over the past decade that directly applies to AI companies and technologies. These lessons can help us avoid making the same mistakes with AI that we did with social media.

In particular, five fundamental attributes of social media have harmed society. AI also has those attributes. Note that they are not intrinsically evil. They are all double-edged swords, with the potential to do either good or ill. The danger comes from who wields the sword, and in what direction it is swung. This has been true for social media, and it will similarly hold true for AI. In both cases, the solution lies in limits on the technology’s use. ...

See the full story here: https://www.technologyreview.com/2024/03/13/1089729/lets-not-make-the-same-mistakes-with-ai-that-we-made-with-social-media/

VR headsets can be hacked with an Inception-style attack

...

A new “inception attack” in virtual reality works in a similar way. Researchers at the University of Chicago exploited a security vulnerability in Meta’s Quest VR system that allows hackers to hijack users’ headsets, steal sensitive information, and—with the help of generative AI—manipulate social interactions. 

The attack hasn’t been used in the wild yet, and the bar to executing it is high, because it requires a hacker to gain access to the VR headset user’s Wi-Fi network. However, it is highly sophisticated and leaves those targeted vulnerable to phishing, scams, and grooming, among other risks.  ...

In the attack, hackers create an app that injects malicious code into the Meta Quest VR system and then launch a clone of the VR system’s home screen and apps that looks identical to the user’s original screen. Once inside, attackers can see, record, and modify everything the person does with the headset. That includes tracking voice, gestures, keystrokes, browsing activity, and even the user’s social interactions. The attacker can even change the content of a user’s messages to other people. The research, which was shared with MIT Technology Review exclusively, is yet to be peer reviewed. ...

See the full story here: https://www.technologyreview.com/2024/03/11/1089686/hack-vr-headsets-inception/?fbclid=IwAR0Z7XV3-LG30e3C5VBKLf5ERhItZgSu0CSyw8Eomb_RXU4ZtXHY6Vx5rjo

iFLYTEK, Zense Entertainment, and Chulalongkorn University Sign Memorandum of Understanding to Accelerate the Advancement of Thai Speech Recognition

... Through this partnership, the aim is to propel the development of cutting-edge Thai speech recognition technology to cover not only its multiple regional dialects but also to mark a significant milestone in the advancement of AI-driven language processing systems tailored specifically for the Thai language. The primary objective of this collaboration is to optimize iFLYTEK’s Thai speech recognition engine by incorporating Chulalongkorn University’s linguistic insights and research. This collaboration will enhance the accuracy, fluency, and usability of Thai speech recognition technology up to 95% plus accuracy. ...This initiative aligns perfectly with iFLYTEK’s commitment to delivering innovative solutions that enable seamless interaction with technology in users’ native languages covering 123 languages. ...

See the full story here: https://temanstartup.com/iflytek-zense-entertainment-and-chulalongkorn-university-sign-memorandum-of-understanding-to-accelerate-the-advancement-of-thai-speech-recognition/

SXSW: Google, J.Crew Execs Discuss Online Shopping In the Age of A.I.

... The two executives are also thinking about how Gen Z will shop with A.I. According to Rincon, the age group of 11-26 sees shopping as a form of entertainment. This demographic group also takes to visual searching, or searching with their phone cameras, instead of typing in keywords. ...

See the full story here: https://observer.com/2024/03/google-jcrew-ai-exec-online-shopping-sxsw-2024/

Employees at Top AI Labs Fear Safety Is an Afterthought, Report Says

..,. One individual at an unspecified AI lab shared worries with the report’s authors that the lab has what the report characterized as a “lax approach to safety” stemming from a desire to not slow down the lab’s work to build more powerful systems. Another individual expressed concern that their lab had insufficient containment measures in place to prevent an AGI from escaping their control, even though the lab believes AGI is a near-term possibility. ...

See the full story here: https://time.com/6898961/ai-labs-safety-concerns-report/

Privacy Is Just No Longer a Thing in Augmented Reality?

Skepticism mounts over Apple’s modest claims on Vision Pro data ...

... Where does that data go, and what could it be used for? The latest wave of interest in augmented reality (AR) makes that a pressing question. But, with trust in technology companies at an all-time low, experts are skeptical tech companies will treat the issue seriously. ...

“I think with virtual reality technologies, because they’re sensing so much of what you do, it necessarily will create troves of data people will want to have access to,” said technology policy lawyer and author Mike Godwin. “Government’s hunger for what you do in cyberspace is already unlimited [...] so, I would expect VR and AR to raise even more questions, though perhaps in the same character of questions as mobile devices already raised.” ...

Apple has stated “privacy is a fundamental human right,” an idea repeated in the company’s Apple Vision Pro Privacy Overview, which it released last month. The white paper detailed efforts to keep user data private through on-device processing and data minimization. Some information the headset collects, such as eye-tracking data, is processed on-device and not shared with Apple or third-party apps. 

Yet there’s caveats, loopholes, and possible end-runs. Apps don’t need to access hand or head-tracking data to function, but they might access that data in certain situations. And while nuanced eye tracking data isn’t shared, the gaze-driven design of VisionOS can’t fully eliminate the problem, as any button a user selects implies where the user is looking. ...

Handing over that data would seem to imply trust—and Godwin finds that difficult to accept. “I think it’s the wrong approach to say we should trust [companies] to do the right thing, because their incentives do not align with consumers.” 

He points out that user data, once collected and stored, is often accessed in ways the user didn’t intend. ...

“Today, even companies that are more trusted in the tech space, are still trusted less because of other companies that have acted badly,” he said. “The only way to rebuild trust is to proactively do things that build trust.”

See the full story here: https://spectrum.ieee.org/apple-vision-pro-privacy

The Hottest Trends in Immersive Environments for 2024

...

1. Immersive Environments Transform Offices ...

2. Hardware for Immersive Experiences Evolves ...

3. The Interaction of AR, VR, and Spatial Computing ...

4. Immersive Environments Grow More Flexible ...

5. Generative AI Enhances Spaces ...

6. Increasing Focus on User Safety and Security ...

See the full story here: https://www.xrtoday.com/mixed-reality/the-hottest-trends-in-immersive-environments-for-2024/

Nvidia’s Jensen Huang: The incredible future of AI

... When asked his advice for Stanford students aspiring to be successful entrepreneurs, Huang talked about the importance of low expectations and high resilience. Greatness, he said, comes from smart people who have suffered from setbacks. This is why, at Nvidia, he talks openly about pain and suffering “with great glee.”

“For all of you Stanford students,” he said, “I wish upon you ample doses of pain and suffering.”

See the full story here: https://siepr.stanford.edu/news/nvidias-jensen-huang-incredible-future-ai

DRONES ARE TRANSFORMING THE BATTLEFIELD IN UKRAINE BUT IN AN EVOLUTIONARY FASHION

See the article here: warontherocks.com/2024/03/drones-are-transforming-the-battlefield-in-ukraine-but-in-an-evolutionary-fashion/

AI Prompt Engineering Is Dead 

Long live AI prompt engineering

... Battle and Gollapudi decided to systematically test how different prompt engineering strategies impact an LLM’s ability to solve grade school math questions. They tested three different open source language models with 60 different prompt combinations each. What they found was a surprising lack of consistency. Even chain-of-thought prompting sometimes helped and other times hurt performance. “The only real trend may be no trend,” they write....

There is an alternative to the trial-and-error style prompt engineering that yielded such inconsistent results: Ask the language model to devise its own optimal prompt.  ...

Battle says that optimizing the prompts algorithmically fundamentally makes sense given what language models really are—models. ...

Image generation algorithms can benefit from automatically generated prompts as well. ...

See the full story here: https://spectrum.ieee.org/prompt-engineering-is-dead