~Spot the differences: Original drug bust photo (left) vs. AI-modified version (right)~

A Maine police department's attempt to add their logo to a drug bust photo backfired spectacularly when citizens caught them lying about using artificial intelligence to alter evidence photos.

"This is NOT an AI-generated photo," Westbrook Police declared on Facebook when first questioned about oddities in their photos of seized meth and fentanyl. They doubled down, insisting "Westbrook PD is not and would never generate an AI photo to try and depict evidence." But internet sleuths weren't buying it. Missing "cookie" stickers, vanished drug residue, and mysteriously altered colors pointed to AI manipulation. Within days, the department was forced to admit the truth — they had indeed used AI software, supposedly just to add their department patch to the image.

The incident exposed an alarming gap between police technological capabilities and understanding. When an officer fed the evidence photo through what they thought was "a photoshop app," the AI rewrote reality — altering drug packaging, removing key details, and creating what amounts to falsified evidence documentation. Even more troubling, the department didn't notice these changes before posting.

"The fact that the person who posted it and put it through ChatGPT didn't notice the differences because they were very obvious…it makes me wonder how much people understand about technology and how easy it is to fool people," said local resident Jessica Wellman, capturing widespread concern about police departments wielding powerful AI tools they don't fully grasp.

::: spoiler Supplementary

Source

:::

Republished from Boing Boing under Creative Commons License.

Extra.

LOL in a lot of ways.

2025 is gonna be interesting.

A new era of Stack Overflow - Stack Overflow

^{stackoverflow.blog}

The photo is from Firefox beta for Android.

The photo is from Firefox beta for Android.

A robot trained on videos of surgeries performed a lengthy phase of a gallbladder removal without human help. The robot operated for the first time on a lifelike patient, and during the operation, responded to and learned from voice commands from the team—like a novice surgeon working with a mentor.

The robot performed unflappably across trials and with the expertise of a skilled human surgeon, even during unexpected scenarios typical in real life medical emergencies.

Robot performs first realistic surgery without human help

In work led by Johns Hopkins researchers, the robot performed unflappably across trials and with the expertise of a skilled human surgeon, even during unexpected scenarios typical in real life medical emergencies

^{The Hub}

People with disabilities, those living in poverty or who have serious health conditions are being left in a bureaucratic limbo due to digital exclusion caused by the Department of Work and Pensions’ (DWP) unchecked roll-out of technologies, Amnesty International said today in a new report.

The report, “Too Much Technology, Not Enough empathy” exposes how the DWP’s constant testing, rolling out, and rolling back of costly artificial intelligence (AI) and digital technologies for Universal Credit (UC), Personal Independence Payment (PIP), and other social security schemes has created an inaccessible social security system for people who are already marginalized and at risk of poverty in the UK.

Many people requiring social security do not have access to digital technologies, internet or internet connected devices. Their affordability coupled with language barriers and long waiting times for telephone services have led to digital exclusion from DWP’s systems.

UK: Government’s unchecked use of tech and AI systems leading to exclusion of people with disabilities and other marginalized groups  - Amnesty International

People with disabilities, those living in poverty or who have serious health conditions are being left in a bureaucratic limbo due to digital exclusion caused by the Department of Work and Pensions’ (DWP) unchecked roll-out of technologies, Amnesty I…

^{Amnesty International}

Source.

A billboard truck spent Wednesday circling the FTC, telling them to focus on Big Tech, grocery prices, and the dangers of privacy-destroying data brokers—instead of holding an out-of-scope workshop on gender-affirming care.

After two incredible years, I’ve decided to step down as CEO of 𝕏.

When
@elonmusk
and I first spoke of his vision for X, I knew it would be the opportunity of a lifetime to carry out the extraordinary mission of this company. I’m immensely grateful to him for entrusting me with the responsibility of protecting free speech, turning the company around, and transforming X into the Everything App.

I’m incredibly proud of the X team - the historic business turn around we have accomplished together has been nothing short of remarkable.

We started with the critical early work necessary to prioritize the safety of our users—especially children, and to restore advertiser confidence. This team has worked relentlessly from groundbreaking innovations like Community Notes, and, soon, X Money to bringing the most iconic voices and content to the platform. Now, the best is yet to come as X enters a new chapter with
@xai
.

X is truly a digital town square for all voices and the world’s most powerful culture signal. We couldn’t have achieved that without the support of our users, business partners, and the most innovative team in the world.

I’ll be cheering you all on as you continue to change the world.

As always, I’ll see you on 𝕏

Source.

https://pirg.org/edfund/updates/court-blocks-rule-to-make-subscriptions-easier-to-cancel/

https://fransskarman.com/phd_thesis_in_typst.html

ETH Zurich and EPFL will release a large language model (LLM) developed on public infrastructure. Trained on the “Alps” supercomputer at the Swiss National Supercomputing Centre (CSCS), the new LLM marks a milestone in open-source AI and multilingual excellence.

• In late summer 2025, a publicly developed large language model (LLM) will be released — co-created by researchers at EPFL, ETH Zurich, and the Swiss National Supercomputing Centre (CSCS).
• This LLM will be fully open: This openness is designed to support broad adoption and foster innovation across science, society, and industry.
• A defining feature of the model is its multilingual fluency in over 1,000 languages.

A language model built for the public good

This summer, EPFL and ETH Zurich will release a large language model (LLM) developed on public infrastructure.

^actu.epfl.ch

ETH Zurich and EPFL will release a large language model (LLM) developed on public infrastructure. Trained on the “Alps” supercomputer at the Swiss National Supercomputing Centre (CSCS), the new LLM marks a milestone in open-source AI and multilingual excellence.

• In late summer 2025, a publicly developed large language model (LLM) will be released — co-created by researchers at EPFL, ETH Zurich, and the Swiss National Supercomputing Centre (CSCS).
• This LLM will be fully open: This openness is designed to support broad adoption and foster innovation across science, society, and industry.
• A defining feature of the model is its multilingual fluency in over 1,000 languages.

A language model built for the public good

This summer, EPFL and ETH Zurich will release a large language model (LLM) developed on public infrastructure.

^actu.epfl.ch