Slopsquatting, Datacentres, and AI’s New Normal: This AI News Roundup

In a week when AI is both enabling new efficiencies and exposing fresh risk, the headlines braid together coding, climate, culture, and the workplace of tomorrow. Slopsquatting, a new AI-driven supply-chain threat, is pushing developers to rethink how they verify dependencies. Datacentres remain a climate pressure point as the biggest cloud players chase growth, and even a glossy AI actor controversy underscores how entertainment and technology are now intertwined. Taken together, these stories sketch a future where the benefits of AI arrive with a larger security, privacy and governance footprint.

Slopsquatting is a class of attack born from AI hallucinations. Large language models generate plausible-sounding, yet fictitious, package names. Attackers register those names and slip malware into a developer’s workflow before the code is even committed. Traditional typosquatting—misspelled libraries—has protections in registries, but AI-generated names aren’t simple misspellings: they are plausible-sounding fan-out that registries can overlook. The risk is real: if a fake package named by the model like ‘crossenv’ emerges as an actual registry item and is loaded into a project, the malware rides in with legitimate-looking dependencies. To counter this, teams should implement automated checks that validate package names against official registries, monitor for unusual package installations, and keep threat intelligence up to date.

On the environmental front, datacenters continue to drive up tech emissions. In the year ending March 2026, Microsoft, Amazon and Google reported emissions totaling 119 million metric tons of CO2e — roughly a third of France’s annual footprint. The construction boom behind those numbers weighs on climate targets, even as these giants push for net-zero ambitions. The takeaway for readers is not doom but due diligence: scale in cloud infrastructure demands governance, energy efficiency, and transparent reporting so that the AI tools powering software and services don’t outrun the planet’s ability to breathe.

Culture and AI intersect in the most public way when entertainment taps synthetic actors. The Guardian’s take on Tilly Norwood—a fully digital ‘AI actor’—highlights the tension between novelty and humanity. A film in development promises a “coming-of-age” story in a world where a cartoon avatar performs alongside humans. Critics question whether a digital entity can inhabit life’s emotional textures or if the industry is chasing novelty at the expense of authentic storytelling. The anecdote is not just about one project; it’s a bellwether for how audiences, studios and AI vendors will negotiate authenticity, consent, and creative control in the years to come.

Beyond screens, professionals are asked to imagine a future where AI support is pervasive but not prohibitive. A Guardian roundtable on AI-proofing careers suggests that many roles—from teaching to law—will still rely on uniquely human capabilities: empathy, complex judgment, nuanced negotiation, and ethical oversight. The data aligns with broader industry sentiment: while AI accelerates routine tasks, people who cultivate critical thinking, governance, and cross-disciplinary skills will likely thrive. In practice, this means updating education and corporate training, as well as building career strategies that emphasize verification, collaboration with AI as a cognitive partner, and a healthy skepticism of “plug-and-play” AI solutions.

1. Forget typosquatting; slopsquatting is the software supply chain threat created by AI coding tools
2. Datacentres drive up big tech’s carbon emissions to a third of those of France
3. AI ‘actor’ Tilly Norwood has a movie coming out. Spare us this future | Dave Schilling
4. Safe from AI: which jobs will help you thrive in the future?
5. Meta ditches Muse Image AI feature because it ‘misses the mark’ on users’ privacy