This belief is pervasive but not backed by evidence.

“Look, the bottom line is Big Tech has been preying on and victimizing our children for way, way too long.” - state Sen. Ben Kramer
apnews.com/article/online…

There are many reforms Congress could make to create a more nimble, productive, and talented federal workforce.

Cutting telework is not one of them.

If workers aren't productive, agencies should dismiss them, not make it harder to recruit top talent.

govexec.com/workforce/2024…

📣 WEBINAR: What do Americans really think about #AI ? Join the discussion on Capitol Hill (Rayburn 2060) on May 21. Speakers include Vinous Ali, Jonathan Dupont, Lee Rainie, Alec Tyson & Daniel Castro. RSVP now! datainnovation.org/2024/05/insigh…

'Because this kind of deepfake detector is driven by probabilities, it can never be perfect.'

Instead of focusing on detecting deepfakes, we should focus on authenticating real content, i.e. creating more trust signals in the information environment.

nytimes.com/2024/05/07/tec…

'The Act does not articulate any threat posed by TikTok nor explain why TikTok should be excluded from evaluation under the standards Congress concurrently imposed on every other platform.'

sf16-va.tiktokcdn.com/obj/eden-va2/h…

Getting facial recognition into airports for the most basic function—having a computer compare a government ID to a face instead of a human—has been a major *uphill* battle.

Yet those advocating for a ban/freeze call this a 'slippery slope.'

nextgov.com/emerging-tech/…

Really interesting.

'In this paper, we introduce a pioneering learning paradigm, non-fine-tunable learning, which prevents the pre-trained model from being fine-tuned to indecent tasks while preserving its performance on the original task.'

arxiv.org/abs/2404.12699

Great piece on the shift of sentiment around the dangers of large models, and the risks of premature regulations.

datainnovation.org/2024/04/the-sh…

Instead of promoting AI in every vehicle, we have Congress pushing the 'AM Radio for Every Vehicle Act.'

A great example of the 'pacing problem' (where tech moves faster than policy).

collins.senate.gov/newsroom/colli…

'The about-face the scientific, academic, and tech communities have made on the risks of large language models (LLMs) creating biothreats should serve as a stark reminder for policymakers about the pitfalls of premature regulation.' - Martin Makaryan

datainnovation.org/2024/04/the-sh…

'reporters love to catch big companies doing something illegal, unethical, or anticompetitive.'

Yep...it's amazing how many tech stories are about 'scandals'.

'crashes involving errors by human drivers are the frequent subject of viral videos in China, feeding a popular perception that self-driving cars may be safer'
nytimes.com/2024/04/29/bus…

'Shein will have to implement include diligent surveillance of illegal products, enhancing consumer protection measures, and ensuring transparency and accountability through audits and reporting requirements.'

euractiv.com/section/platfo…

'TikTok accounts for a small share of ByteDance's total revenues and daily active users, so the parent would rather have the app shut down in the U.S. in a worst case scenario than sell it to a potential American buyer'
reuters.com/technology/byt…

The number of CAPTCHAs I have failed makes me question whether I am really human.

Using AI to understand how insects fly (and testing those ideas on a tiny robot fly). Some fascinating research:

arstechnica.com/science/2024/0…

Banning TikTok is bad policy, plain and simple. Information Technology and Innovation Foundation statement from Daniel Castro ⤵️
itif.org/publications/2…

China ordering Apple to block WhatsApp and Threads in the hours before Congress votes on a TikTok ban is incredibly poor timing.

telegraph.co.uk/business/2024/…