SentinelOne says macOS.Gaslight uses prompt injection to mislead AI-based malware analysis, steal data, and use Telegram for C2.
The post North Korea-Linked macOS Malware Uses Prompt Injection to Evade AI Analysis appeared first on TechRepublic.

SentinelOne says macOS.Gaslight uses prompt injection to mislead AI-based malware analysis, steal data, and use Telegram for C2.
The post North Korea-Linked macOS Malware Uses Prompt Injection to Evade AI Analysis appeared first on TechRepublic.

AI has given cybercriminals a big advantage in attacking organizations, which they are using to go after weaknesses on edge devices
The post AI-Driven Identity Attacks Are Surging, PwC Warns appeared first on TechRepublic.

The ban on ozone-depleting substances that successfully reversed the growth of the hole in the ozone layer isn’t seen as a missed opportunity. On the contrary, the quick global response is one of the best cases of common-sense environmental action. But what if it could have been done even earlier?
The fact that chlorofluorocarbons (CFCs)—chemicals once common in aerosol cans and refrigerant loops—could destroy ozone in the atmosphere was discovered in 1974. Within just a few years, bans on CFCs began to roll out based on the projected consequences. The seasonal ozone “hole” discovered over Antarctica in 1985 pushed things along even faster, and in 1987 an international agreement was signed to phase out CFCs everywhere.
A new study led by Jian Guan at MIT asks an interesting what-if question: Would it have been possible to detect this problem even sooner with today’s scientific tools?

Sony recently informed its PlayStation customers in the United Kingdom that they will no longer be able to watch previously purchased movies and shows from production and distribution company StudioCanal. As of September 1, affected customers will no longer be able to stream 551 titles from the PlayStation Store.
In a legal notice first spotted by gaming news outlet PlayStation LifeStyle, Sony said that affected customers will lose the ability to stream titles including Outrage: Way of the Yakuza, Paddington, Paddington 2, Pan’s Labyrinth, Rambo 3, Terminator 2: Judgment Day, and The Boy in the Striped Pajamas “due to our content licensing agreements.” As of September, Sony will remove any affected titles that UK users bought from their PlayStation library, per the notice.
It’s possible that Sony may still make a deal with StudioCanal by September 1, or even after, that would allow users to keep watching the content they bought. This happened in 2023, when Sony said it would have to pull 1,318 seasons of Discovery shows from customers’ libraries. A few weeks after its announcement, Sony said that it would not pull the content because it had updated its licensing arrangements with Discovery.

Microsoft’s biggest hardware releases of 2026 include new Surface Laptop and Surface Pro business models, lower-cost Surface configurations, and Project Solara.
The post Microsoft’s Biggest Hardware Releases of 2026 So Far: Surface Laptop, Surface Pro, and More appeared first on TechRepublic.

The Fourth Amendment protects a user’s “location history,” the Supreme Court ruled Monday.
The same logic already applied to a cellphone’s tracking, and the high court found “no good reason exists to reach a different result for Location History” collected by third parties like Google.
Split 6-3, the majority agreed that the government needs a warrant and must show reasonable cause to turn a phone’s location-tracking services into a government surveillance tool.

When last we looked at the state of the US grid, the ongoing explosion in solar energy had turned it into a major contributor, but one that still lagged well behind fossil-fuel-powered generation. So it was a bit of a surprise when preliminary data suggested that May 2026 saw solar power pass coal-fired generation for the first time in the US. Now, with the official release of April grid data by the Energy Information Administration, we can see that production of solar electricity had passed coal a month earlier—with a bit of a caveat.
The caveat being that a substantial chunk of that solar production never reached the grid, since it’s produced by rooftop installations and used in the building they sit atop.
The situation heading into April/May was pretty simple. After a brief resurgence last year, coal use resumed its decline, despite repeated government attempts to prop it up. Meanwhile, solar continued its rapid growth, driven by its position as the cheapest way to add generating capacity in most of the US. But this growth started from a small base, and the early months of the year are marked by seasonally low solar production. As a result, growth above 20 percent year over year still left solar providing only 6 percent of the power on the US grid, a sharp contrast to coal’s 16 percent.

Gender-based violence researchers, feminists and activists use the term “rape culture” to describe a society where violence against women, girls and gender-diverse people is normalized, victims are blamed and perpetrators are routinely excused.
Earlier this month, it reared its head again: this time among school-aged youth in an Alberta school.
Boys at Camrose Composite High School reportedly listed girls they wanted to sexually assault or mutilate in a series of Snapchat group messages. One boy reportedly called himself a “rapist” in the chat.
Following this discovery, students walked out in protest against this technology-facilitated gender-based violence.
Technology-facilitated gender-based violence involves the use of digital and emerging technologies to stalk, surveil, harass or humiliate a target — typically a woman, girl or member of an additionally equity-denied group.
Instances of this form of violence are not uncommon, including among youth. In fact, within the same week, two 14-year-old boys were charged after they allegedly used artificial intelligence to create graphic deepfake nudes of multiple girls from Edmonton high schools.
As of publication, no charges or school disciplinary findings have been publicly confirmed in connection with the Camrose incident. But what happened in the days after the chat surfaced says as much about Canada’s rape culture as the chat itself.
The problem isn’t just boys engaging in technology-facilitated gender-based violence. It’s also how others in their environment react to these kinds of events.
First, there is the culture of impunity. The social media threads were allegedly active since fall 2025, with multiple boys said to have contributed over that period of time. Then, according to a Grade 12 student who was part of the protest, when other students tried to report the chat to school officials, they were told by a staff member that “boys will be boys.”
The phrase “boys will be boys” is a hallmark of rape culture. This saying deftly writes off responsibility for the boys and men who perpetuate abuse. It also leaves girls, women and gender-diverse individuals to deal with the consequences without benefiting from a sense of justice, closure or, at bare minimum, support.
Consequences of gender-based violence can be physical, emotional, sexual or economic, to name a few. For example, costs can quickly mount when therapy is needed or a survivor misses work. Meantime, a culture that normalizes gender-based violence creates a space where abuse continues unchecked.
Equally notable is the school’s persistently gender-neutral language. Instead of disclosing that the thread’s creators were boys, the Battle River School Board’s statements call them “students.” Such gender neutrality rewrites Canada’s gender-based violence problem as something all students face or perpetrate equally, which is far from the case.
This gender-neutral language had downstream effects. It made it hard for media outlets, in the wake of the Camrose events, to clarify to the public that gendered abuse specifically had taken place.
As a scholar whose work explores technology-facilitated gender-based violence and who was interviewed by news media in the aftermath of the Camrose events, I can attest to this. The school board’s and police’s gender neutrality made it hard to comment on the case in a way that meaningfully raised awareness about violence against girls or allowed for an intervention in the issue.
The use of social media apps and AI tools by school-aged children (typically boys) to create deepfake pornography of classmates (typically girls) is on the rise.
Research has linked growing misogynistic attitudes among teen boys broadly to influencers like Andrew Tate and manosphere rhetoric — a trend made worse by a society that ineffectually addresses gender-based violence.
This degendering of violence — leaving out specifics about who most perpetrates and who most experiences it — is common in rape culture. And it makes it harder for critics, scholars and activists to paint a complete picture of Canada’s deadly problem.
While violent crime has decreased overall through the years in Canada, gender-based violence has continued to rise.
One woman or girl is killed every 48 hours, primarily by men. One in four women reports gender-based violence to the police, although experts believe one in three actually face it, but are too afraid, due to stigma or potential disbelief, to report. Women are five times more likely to face sexual assault than men, and four times more likely to experience intimate partner violence.
The risk, however, is distributed in additionally unequal ways. The highest rates of gender-based violence are experienced by specific marginalized groups.
Indigenous women and girls are far more likely to be murdered or to disappear than other women in Canada. Transgender and gender-diverse people experience violent victimization at substantially higher rates than cisgender people. Women of colour, and those with disabilities or from rural areas, are among the groups facing the highest rates of GBV.
Still, there are hopeful aspects to the Camrose case.
Young people have a voice, as evidenced by the student protests. They are making clear that they are experiencing Canada’s gender-based violence problem, especially as it extends onto digital platforms, and that this is not OK.
It is also heartening to know that boys were standing alongside girls and gender-diverse individuals near their high school in support, and to hear their thoughtful commentary in news broadcasts. Hopefully, more will join.
Similarly, in the case of the two boys who created deepfakes, a schoolteacher reported the matter and charges were laid. When educators act meaningfully to protect students, it is another step in the right direction.
We can all learn to better respond to gender-based violence, both online and offline. First, don’t normalize misogynistic threats and behaviour — it extends the harm that victims will experience while minimizing responsibility for the perpetrator.
Second, don’t degender the crime. Doing so makes this deadly Canadian issue seem like something all people face equally, when it isn’t. Third, when students are courageous enough to come forward about gender-based violence, acknowledge their experience and be of assistance to them. Research shows harm is reduced when survivors experience support and care.
And finally, help men and boys by providing a safe space for them to be vulnerable and express emotions without shame. They too have been harmed by a society that demands that they be tough, stoic and aggressive and that they reject what is “feminine” as weak and less than.
![]()
Nicolette Little receives funding from the Social Sciences and Humanities Research Council of Canada.

The MIT Music Technology and Computation (MTC) Graduate Program — launched in fall 2024 as a collaboration between the Music and Theater Arts Section in the School of Humanities, Arts, and Social Sciences (SHASS), and the School of Engineering (SoE) — presented its inaugural MIT Music Technology Research Showcase on May 13. The event played to a standing room-only house in the Edward and Joyce Linde Music Building’s Thomas Tull Concert Hall and featured diverse and captivating research presentations and music performances.
The celebratory occasion featured MTC’s first five enrollees (all of whom were previously MIT undergraduates), alongside several PhD students and faculty. Each scholar presented inspiring exemplars of artful engineering that reflected the broader and burgeoning music technology scene at MIT.
The 90-minute event exhibited a broad array of research projects, including a real-time visualization of what an AI co-improvising agent is about to play on a piano; a sound-art installation based on noisy network communication; a hip-hop dance circle where music is generated from dancing; and the use of electroencephalogram (EEG) signals to identify the musical tunes that our brains are imagining.
“A new space for exploration and insights”
An interplay of technical presentation with live performance, the showcase began with remarks from SHASS Dean and professor of philosophy Agustín Rayo, SOE Dean and professor of chemical engineering Paula Hammond, and MTC Director and professor of the practice of music Eran Egozy.
Rayo began, “The goal of this program is simple — for MIT to lead the world in music technology theory and application,” adding “it’s not just about making music with technology; it’s also about working across disciplines to help better shape the future of expression in an AI-driven world, all while reflecting MIT at its best.”
Rayo noted the graduate program was made possible in part by the opening of the Edward and Joyce Linde Music Building in 2025, which provided new classrooms, studios, rehearsal spaces, and a dedicated music technology lab. He also credited the MIT Schwarzman College of Computing for its support for the graduate program.
Hammond followed: “As those in this room already know, music and engineering share some common roots. Both rely on mathematical precision and are informed by defined structures, rhythms, and frequencies. Both demand hard work and technical know-how, paired with inspiration and imagination, to create something entirely new. Given those congruities, it’s no surprise that so many faculty, students, and staff members across MIT are also accomplished musicians and artists.”
She continued, “Our music program is a gem. Only at MIT could we bring the top technologists and the top musicians together to create unique opportunities for collaboration. Here we have brought together faculty and students who identify strongly with both music and engineering to form a new space for exploration and insights. It’s a strong example of the collaborative culture that defines the Institute.”
Egozy called the event a “harmonious hybrid of concert and symposium,” and recollected, “it’s a little mind-boggling to see what our students have achieved in just one short and fast-paced year. While we originally debated the trade-offs between a one-year and a two-year master’s program, I think this cohort really showed us that we can make huge strides in learning and research abilities in a concentrated period of time.”
Student research on display
One of those students is Claire Southard ’25, SM ’26, who developed a machine-learning model used to identify musical notes hidden in EEG signals.
Southard explains, “every year, musicians are diagnosed with movement disorders such as Parkinson’s disease and dystonia, or experience injuries that prevent them from controlling their hands and bodies in the ways required to play their instruments. Because of this, too many musicians are forced to stop doing what they love. My work explores one strategy to help such musicians perform again by translating the music they’re trying to play directly from their brain activity — bypassing the need for motor control altogether. To do this, I trained machine-learning models to predict the music a person is imagining from their brain activity measured using EEG, and many of the predicted pieces were found to be recognizable representations of what the user imagined. By designing a system that allows musicians to create music regardless of their physical abilities, I hope this work helps bring a more accessible future for music performance closer to reality.”
Before joining MTC, Southard was initially unaware of the breadth, scope, and magnitude of what the program could offer for further pursuing and realizing her interests. “The MIT Music Technology and Computation Graduate Program taught me so much about the possibilities at the intersection of STEM and the arts,” she says. “When I first started the program, I honestly wasn’t sure what counted as ‘music technology.’ Through classes, research, and conversations with faculty, guest speakers, and peers, I learned the field was far broader and more fascinating than I could have previously imagined.”
She continues, “coming from a background in neuro- and computer science, many of my undergraduate projects happened entirely on devices. But this program allowed me to encounter more hands-on experiences, from conducting audio recordings to building electronic musical instruments from scratch.”
Another MTC graduate, and student speaker at the 2026 SHASS Advanced Degree Ceremony, Mariano Salcedo ’25, SM ’26, presented a custom web application allowing anyone to create unique emergent visuals that are driven by real-time streaming music. To accomplish this effect, Salcedo built algorithms that leverage the complex visual behavior of self-organized systems as the means toward an aesthetically synergetic end.
In his Advanced Degree Ceremony oration, Salcedo expressed his gratitude and admiration for the passionate people that he’s met not only in MTC, but at MIT overall. In an appropriately compassionate mode, he empathetically opined, “I think what times like this call from us is to lead the way in human and humane-centered technology, which means we don’t only just ask what we can build, but we also ask who is it going to affect, who is not going to affect? Who does it benefit?”
Music technology thriving at MIT
Associate Professor Anna Huang SM ’08 of MTA and the Department of Electrical Engineering and Computer Science (EECS, through SCC), a graduate of the MIT Media Lab, and one of the world’s leading researchers in collaborative human-AI music-making, echoed both Southard and Salcedo’s sentiments through her keynote presentation, “In Search of Resonance in Human-AI Interaction.” A compelling and intimately conversational address, her speech emphasized the importance of centering the human musician in all that is done relating to AI, while also making efforts to include all musics of the world in its discourse at every opportunity.
With many of her family members in the audience, Huang reflected, “I have the privilege of being in both MIT Music and EECS — an interdisciplinary, shared space. What does it mean to build music technology in this context? We’re surrounded by extremely talented musicians, so we take this co-design approach: We work with these musicians, we go into the studio, and every week we try something. And the technology grows with the creative process. We’re always trying to push both of these forward, and it’s always on the edge. It’s very, very rewarding. It’s where I feel most at home.”
Huang also explained how this practice sets the stage for a new Studies in Music Technology subject that she will be co-teaching in the fall with recently appointed Professor of Theater Arts Grisha Coleman. Class 21M.369/569 (Tuning Attention: Creative Practices in Movement, Sound, and AI) proposes that the study of sound and movement practices can inform how we build and envision computational systems, focusing particularly on our relationship to AIs. It will introduce students to a range of musical practices in improvisation and somatics by way of motion-capture technologies, critical interaction design, generative modeling, and algorithms for interpretability and learning through human feedback.
All considered, the future of the MIT Music Technology and Computation Graduate Program is bright. Egozy says MTC admitted 10 master’s students for the 2026-27 academic year from over 100 applicants. Unlike this year’s class, next year’s students will not only include recent MIT undergraduate alumni, but also new faces to campus.
“Widening the pool to graduates of other schools and institutions will bring an extraordinary wealth of perspectives and experiences to the program. Additionally, all three shared faculty between MTA and EECS — including Mark Rau, Paris Smaragdis SM ’97, PhD ’01, and Huang — are inviting new Music Technology PhD students to their labs by way of EECS,” Egozy says.
Embodying its mission, MTC is proving to be a vibrant, multidisciplinary program that attracts many kinds of students with a variety of career objectives from wide-ranging backgrounds.
“Despite their diversity, our students all possess a central commonality,” Egozy says, “not just a shared love for music, but also a deep desire to augment that passion by way of technology in a very warmhearted, humanitarian way.”
List of projects
Rachel Loh, Quanta Fellow in Music Technology and Computation: “Visualizing the Internal State of Music Models for Live Human-AI Improvisation”
Noble Harasha, Quanta Fellow in Music Technology and Computation: “Modeling Subjectivity and Collective Sensory Perception as Noisy, Analog Communication in Feedback-Driven Networks”
Z Chen, Quanta Fellow in Music Technology and Computation: “Generative Music as a Catalyst for Social Choreography”
Nithya Shikarpur: “The Moving Drone: A Live Improvisation in the Context of Hindustani Music with the Human Voice, Generative models, and Loops”
Mariano Salcedo, Alex Rigopulos (1992) Fellow in Music Technology and Computation: “Neural Cellular Automata for Interactive Music Visualization”
Claire Southard, John Piscitello Fellow in Music Technology and Computation: “Neural Decoding of Imagined Music”
Stephen Brade, Suwan Kim, Valerie Chen: “Whale, Cello (there?): A Musical Dialog between Cello and a Real-time Diffusion Model Trained on Whale Songs”


Music streaming service Tidal announced it won’t pay royalties for AI-generated music in an email to users and an announcement on its website published Monday. “Tidal’s priority is ensuring royalties go to original works directly produced, written, and performed by people,” the announcement reads. “We will therefore not knowingly attribute royalties to music we identify as wholly AI-generated.”
Like much of the internet, music streaming services are awash in AI-generated slop. Spotify promised to fight AI spam with labeling and filtering but also embraced the broader trend of AI music. AI-generated bands like The Velvet Sundown and Breaking Rust have millions of listens on Spotify and make the service money. In May, Spotify announced a deal with Universal that would let fans create “covers and remixes of their favorite songs.”Soon Spotify customers will be able to push a button and discover what Metallica would sound like if it were a reggae band.
Tidal is trying something different. The streaming service isn’t a giant in the field — Apple Music, YouTube, and Spotify dominate the charts — but it’s built a reputation by collaborating with artists, giving them a bigger cut of the streaming profits, and focusing on delivering high quality versions of audio. Tidal is the streaming service for listeners obsessed with bit-rate and FLAC. It’s for people who have $200 digital-to-analog converters next to their computer.
The company said it won’t pay for “wholly” AI-generated music but it also said it won’t remove AI-tainted music from the platform entirely. Like Spotify before it, Tidal said it’s going to work to identify the AI slop on its platform, label it, and hold AI-generated music to a “higher standard of content integrity.” Spotify said something similar last year, but there are still plenty of unlabeled AI-generated tracks on the platform.
Tidal also said it won’t remove AI-tainted music entirely. “Artists should have the freedom to create with AI tools, and listeners should have the autonomy to choose the type of content they consume,” it said. As of this writing, The Velvet Sundown and Breaking Rust are both live on Tidal. Breaking Rust’s bio identified it as AI-generated country music, but The Velvet Sundown had no bio at all.
“Tidal will not allow music that is 100% AI-generated to be monetized. No royalties will go to such releases, nor will AI-generated uploads be eligible for direct-to-fan sales,” the company said in an email to its users.
It elaborated on its website. “Starting today, AI-generated music will not be monetizable,” it said. “We are only in the beginning of the era of AI-generated music. We acknowledge the ongoing debate regarding whether certain AI-generated music (e.g. AI-generated music developed from fairly and properly licensed models) should be entitled to earn royalties. This debate will continue as the technology advances and rightsholders and AI music platforms develop licensing models.”
It’s unclear if The Velvet Sundown and other bands like it will keep making money on Tidal. These bands are black boxes. Are they wholly AI-generated and therefore exempted from compensation from Tidal? Or are they human-directed enough that the streaming service will continue to compensate them. 404 Media reached out to Tidal to ask but did not hear back.
On June 28, the day before Tidal’s announcement, The Velvet Sundown released a cover of Dolly Parton’s “I Will Always Love You” on Spotify and Tidal. It’s atrocious and it’s not labeled as AI-generated on either service.
“We exist to confuse music journalists, comfort robots, and help Spotify executives sleep at night,” says the frontpage of The Velvet Sundown’s website. “We were basically built for it, engineered to fill playlists, avoid royalties, and haunt your Discover Weekly like a ghost with good taste. Is it art? Is it a loophole? Either way, it streams beautifully.”