<![CDATA[Why teens use drugs and alcohol: avoidant attachment and emotional isolation fuel addiction; attachment-based therapy rebuilds bonds and cuts relapse.]]>
The Download: energy transmission and US threats against Chinese AI
This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology.
The power line that could reshape New York’s grid is hitting snags
During a heat wave on July 3, New York State’s grid imported enough electricity from Canada to meet about 9% of its total demand that day.
Some of that power shuttled in on a 339-mile power line stretching from Quebec to Queens. It opened in May and is officially the longest underground transmission line in North America. It could provide up to 20% of New York City’s electricity demand, largely with abundant hydropower from Quebec.
One wrinkle: The line has been down for most of this month, and some experts are concerned about how drought will affect the power supply feeding it.
Still, the line could help shape the future of our grid, if it can overcome these sorts of snags. Read our story to understand how.
—Casey Crownhart
This story is from The Spark, our weekly climate tech newsletter. Sign up to receive it in your inbox every Wednesday.
The must-reads
I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology.
1 The US Treasury is threatening to sanction Chinese AI companies
Treasury secretary Scott Bessent has accused Moonshot of improperly distilling Anthropic’s Fable model. (TechCrunch)
+ Nvidia’s Jensen Huang is arguing that America has nothing to fear from Chinese AI. (Axios)
+ Like it or not, Chinese models are now part of the global AI infrastructure. (Rest of World)
+ China’s AI models have Trump’s AI world at war with itself. (MIT Technology Review)
2 Why the OpenAI hack is the scariest AI mishap yet
AI’s capabilities seem to be starting to outpace our current ability to control them. (The Economist $)
+ Hugging Face had to turn to a Chinese AI model to rescue it from the hack. (BI)
3 Visually impaired Europeans can now get an implant that restores sight
And Americans may not have to wait long to receive it, too. (STAT)
+ This retina implant lets people with vision loss do a crossword puzzle. (MIT Technology Review)
4 A bellwether lawsuit suing Meta for social media addiction has been dropped
There are, however, many more waiting in the wings. (NYT $)
5 Here’s how ICE gets its hands on Americans’ data
As soon as you open a credit card or phone account, its agents can see where you live. (404 Media)
+ States are warring with the Trump administration over the right to see ICE agents’ faces. (Wired $)
6 We urgently need to grapple with AI’s environmental impact
As the world warms, is the price we’re paying worth it? (The Verge)
+ We did the math on AI’s energy footprint. (MIT Technology Review)
7 Privacy issues with smart glasses need an industrywide fix
That’s according to Samsung, which is unveiling glasses it developed with Google this fall. (Bloomberg $)
8 The US Army is begging soldiers to limit their AI use
The token crisis comes for us all eventually, it seems. (Ars Technica)
9 Why does lettuce keep making Americans sick? 
It’s pretty simple: a lot of people eat it, and it doesn’t get cooked. (Wired $)
10 Pokemon Go is the perfect game to play this summer
It’s fun, collaborative, and it gets you outdoors. (Guardian)
Quote of the day
“It went off and did this hack all by itself, as far as we can tell. This is the highest level of autonomy that we’ve seen in the use of a large language model for cyber operations.”
—Colin Shea-Blymyer, a cybersecurity research fellow at Georgetown University, tells NPR why the OpenAI hack on Hugging Face is so alarming.
One More Thing
KAGAN MACLEOD
Welcome to the dark side of crypto’s permissionless dream
Jean-Paul Thorbjornsen is a founder of THORChain, a blockchain through which users can swap one cryptocurrency for another and earn fees from making those swaps.
But is he responsible for what it’s used for? It’s a question that matters because in January last year, its users lost more than $200 million in cryptocurrency after THORChain transactions and accounts were frozen by an admin override, which users believed was not supposed to be possible given the decentralized structure. It’s also been used by North Korean hackers to move $1.2 billion of stolen ethereum.
Thorbjornsen explains this all away as a function of THORChain’s decentralized and permissionless nature. Read our story exploring whether we should believe him or not.
—Jessica Klein
We can still have nice things
A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.)
+ A musician developed an ingenious way to strum a guitar with an electric fan.
+ Ukraine’s tunnel of love is a leafy green corridor of romance that’s straight out of a fairy tale.
+ The driver of a giant banana has been pulled over 100s of times, but still won’t ditch his treasured ride.
+ Ever wonder which albums and songs truly stand the test of time? The Greatest Music tries to answer that via an algorithm that analyses hundreds of “best of” lists.
Turning Solar Power Into Protein
According to projections from the United Nations, global food demand could increase by around 60 percent by 2050, while only about two percent additional agricultural land is expected to become available. Researchers at the Technical University of Munich (TUM) report that they are exploring new approaches to safeguard food security. A team at the TUM Campus Straubing has developed a process for producing crucial amino acids from carbon dioxide, hydrogen, and renewable energy.
Viktoria Lehmann, a doctoral candidate at TUM, describes one potential application for biotechnologically produced amino acids.
“A dairy cow needs far more than the grass growing in its pasture. High milk yields require supplemental protein, which is typically supplied through animal feed. These feeds are enriched with amino acids, the chemical building blocks of proteins,” she explains. “Across livestock production systems worldwide, millions of tons of amino acids are used as feed additives. However, their production consumes large amounts of land, water, and other resources. We wanted to find a more resource-efficient way to meet this protein demand.”
In a recently published study “Plug and Play – Enzymatic Amino Acid Production from Methanol and Carbon Dioxide” in Nature Communications, the team demonstrated its approach. The concept behind it: solar energy is converted into electricity using photovoltaic systems. This electricity is used to generate hydrogen, which, together with carbon dioxide, is converted into methanol—an alcohol widely used in industry as a chemical precursor. Specialized enzymes then convert the methanol step by step into amino acids. Which amino acid is produced depends on the specific enzymes used.
“Plants use sunlight to build biomass, but they are relatively inefficient at doing so. We are investigating an alternative pathway in which renewable energy is first converted into chemical energy carriers and then into valuable protein building blocks,” notes Volker Sieber, PhD, professor of chemistry of Biogenic Resources and Rector of the TUM Campus Straubing. “In the long term, this approach could help make more productive use of available land and enable a more sustainable production of amino acids.”
A modular platform technology
In 2023, the researchers demonstrated the production of the amino acid L-alanine from green methanol. Their latest work expands the approach to a total of seven amino acids. “Our modular plug-and-play concept can be compared to a construction kit,” says Vivian Willers, PhD, whose doctoral research laid the foundation for the study. “What started with a single amino acid is increasingly evolving into a platform technology for producing protein building blocks from renewable energy.”
The team successfully produced the amino acids glycine, serine, L-aspartic acid, L-valine, L-glutamic acid, and L-proline. In the future, this technology could help reduce dependence on protein-rich feed ingredients such as soy, which are not always produced sustainably. These amino acids are also key components of nutrient media used in cultured meat production. As a result, the researchers see applications extending well beyond conventional agriculture.
While the team was able to demonstrate the entire process chain—from carbon dioxide via methanol ultimately to amino acids—the current production volumes are still too low for commercial use. The researchers are working to further improve the performance of the enzymes involved.
“Our work is primarily a proof of technological feasibility,” points out Sieber. “We have shown that a broad range of biologically relevant amino acids can be produced from CO₂-based methanol. This opens up new possibilities for the sustainable production of protein building blocks.”
The post Turning Solar Power Into Protein appeared first on GEN – Genetic Engineering and Biotechnology News.
<![CDATA[Top 2026 psychiatry conference takeaways: mixed-feature bipolar care, GLP-1 addiction trials, schizophrenia adherence, TD gaps, AI prodrome screening.]]>
Editorial: Emerging treatment approaches for substance use disorders
Development and validation of the digital well-being scale using university students
IntroductionDigital technologies are central to university students’ academic, social, and personal lives, offering opportunities for learning and connection while also introducing challenges such as distraction, emotional strain, and difficulties with self-regulation. Despite increasing interest in digital well-being, there is a lack of comprehensive, psychometrically sound instruments specifically designed to assess digital well-being among university students. This study developed and validated the Digital Well-Being Scale (DWS) for use in higher education.MethodsA cross-sectional survey was conducted among 1,533 undergraduate and postgraduate students from a public university in Ghana. Scale development involved literature review, expert evaluation, pilot testing, and psychometric validation. Exploratory factor analysis (EFA) was used to identify the underlying factor structure, followed by confirmatory factor analysis (CFA) to evaluate construct validity. Reliability, convergent validity, discriminant validity, and measurement invariance across gender were also assessed.ResultsEFA supported a six-factor solution comprising Task Interference, Digital Safety and Responsible Use, Perceived Control and Satisfaction, Digital Life Balance, Emotional Regulation, and Digital Dependence and Frustration. Following CFA-based refinement, a 41-item scale was retained. The six-factor first-order model demonstrated excellent fit to the data and outperformed a higher-order model. The DWS showed satisfactory internal consistency, convergent validity, and discriminant validity across all dimensions. Measurement invariance analyses supported configural and metric invariance across gender, although scalar invariance was only partially supported.DiscussionThe DWS provides a theoretically grounded and psychometrically robust multidimensional measure of digital well-being among university students. The findings suggest that digital well-being encompasses behavioural, emotional, cognitive, self-regulatory, and responsible dimensions of digital engagement rather than representing a single, unidimensional construct. The scale offers a valuable tool for research, screening, and intervention aimed at understanding and promoting digital well-being and digital mental health in higher education settings.
<![CDATA[Learn how clinicians tackle mood and substance use disorders together, boost patient engagement, and expand evidence-based addiction medications for bipolar and alcohol use.]]>
Physical education performance as a protective pathway: breaking the cycle between learning burnout and gaming disorder in Chinese adolescents
Background and AimsGaming disorder (GD) and learning burnout (LB) are critical issues impacting adolescents, with GD recognized by the World Health Organization as a behavioral addiction and LB contributing to academic disengagement. While prior research has examined bivariate relationships, such as GD’s negative correlation with academic performance and physical education performance, or LB’s association with poor grades, no study has integrated all four variables into one model to explore their dynamic interactions longitudinally.MethodsA cross-lagged panel network (CLPN) model was applied to longitudinal data from 811 Chinese middle school students collected at two time points.ResultsGD and LB showed minimal direct effects on academic performance or physical education performance. Instead, academic performance and physical education performance acted as protective factors, significantly alleviating LB, particularly among boys. These two protective factors were mutually reinforcing. Cognitive exhaustion (a core component of LB) and GD functioned as reciprocal risk factors.ConclusionInterventions for adolescent maladjustment should prioritize addressing learning burnout through academic and physical avenues rather than overemphasizing gaming disorder, with tailored strategies for different genders.
Opinion: MAHA is rewriting the vocabulary of American mental health care
At a May MAHA Institute summit organized around the theme of “overmedicalization,” the health secretary announced an action plan to promote psychiatric deprescribing. At first look, it seemed innocuous. The Substance Abuse and Mental Health Services Administration (SAMHSA) would study prescribing trends and publish fact sheets. Medicare would clarify how clinicians can be paid for the attentive work of tapering a patient off of a medication (which is already a part of routine clinical care). Webinars would teach prevention and “holistic” care. A technical expert panel would convene over the summer to make further recommendations.
In reality, this announcement, and the steady stream of actions over the past 18 months, mark a quiet rewriting of the vocabulary of American mental health care — a massive rhetorical shift enacted while programs and protections that would actually solve the problem are dismantled.
Modular Antibody–ADC Click Chemistry Overcomes Tumor Drug Resistance in Mice
Cancer drugs that can find and kill tumor cells with molecular precision have reshaped oncology—but their power fades when tumors evolve, diversify, or simply stop presenting the right molecular targets. Now, researchers at Washington University (WashU) School of Medicine in St. Louis have unveiled a modular “click‑to‑assemble” strategy that supercharges these therapies inside the body, allowing them to hit multiple tumor targets at once and reverse drug resistance in mice.
The study, published in Nature, is titled “Modular in vivo antibody–ADC click to reverse drug resistance in tumors,” and introduces a bioorthogonal ligation strategy that forms a dual‑targeting antibody–drug conjugate (ADC) with improved tumor uptake and therapeutic efficacy. In pancreatic, gastric, and breast cancer models, the approach slowed or halted tumor progression—even in tumors with low, ultralow, or heterogeneous HER2 expression, where conventional HER2‑directed ADCs typically fail.
“We’ve shown that when two cancer‑targeting antibodies bind together inside the body, they accumulate at the tumor more effectively and improve treatment response,” said senior author Patrícia M. Ribeiro Pereira, PhD, assistant professor of radiology at WashU Medicine’s Mallinckrodt Institute of Radiology. “There is a lot of excitement here because we have shown that it isn’t necessary to create a whole new drug platform for each therapeutic target. We can repurpose antibodies that already exist to improve treatments.”
The team’s strategy—called antibody–ADC click—relies on the rapid ligation between trans‑cyclooctene (TCO) and tetrazine moieties. In the study, one antibody (such as panitumumab, targeting EGFR) was modified with TCO, while a second antibody or ADC (such as trastuzumab‑deruxtecan, T‑DXd) carried tetrazine. When administered sequentially, the two components “clicked” together, forming a higher‑order complex that internalized more efficiently and delivered more cytotoxic payload.
The paper described the problem bluntly: “ADC efficacy remains constrained by its dependence on a single target antigen, which limits tumor targeting and promotes resistance in heterogeneous tumors with variable and low antigen expression.” The click strategy, the authors wrote, “provides a modular and translatable approach for enhanced targeted drug delivery in heterogeneous tumors.”
In mice bearing tumors with mismatched HER2 and EGFR expression, the clicked ADCs accumulated at levels up to 3.2‑fold higher than standard ADCs.
The modularity is the point. “The approach is flexible enough to be adapted to new cancer targets as we learn more about what drives treatment resistance,” Ribeiro Pereira added.
Next steps include further development and optimization of the click-enabled ADC platform prior to future clinical translation, along with expanding the platform to notoriously hard‑to‑treat cancers. “We’re trying to optimize this tool to help antibodies reach tumors that are normally very difficult to treat, such as brain tumors,” Ribeiro Pereira said.
The post Modular Antibody–ADC Click Chemistry Overcomes Tumor Drug Resistance in Mice appeared first on GEN – Genetic Engineering and Biotechnology News.