Hacker Community
The VCs Who Screamed That Biden Would Kill Powerful AI Models Seem Quite Chill About Trump Actually Doing It (Techdirt)
Summary: The Trump administration forced Anthropic to shut down its Fable 5 and Mythos 5 models via export controls, citing national security after Amazon researchers demonstrated a jailbreak. The jailbreak was defensive in nature, according to a cybersecurity expert who reviewed the report. Silicon Valley VCs who previously screamed that Biden would kill AI models are now quiet or supportive of Trump’s far more aggressive action. The move sets a precedent that the government can ban any AI model by simply invoking national security with no due process.
Why it matters: This establishes a dangerous precedent: the US government can now unilaterally shut down AI models by claiming national security, with no evidence or due process, and the tech industry’s most vocal deregulation advocates are cheering it on.
Context: Anthropic had been asking the Trump administration for regulation, and the company was already in a dispute with the DoD over terms of use. Amazon, an Anthropic investor, tipped off the government about the jailbreak.
"The VCs Who Screamed That Biden Would Kill Powerful AI Models Seem Quite Chill About Trump Actually Doing It from the seems-bad dept Late Friday, Anthropic shut down access to its just-released." — TECHDIRT
Commentary: This is a field report from the ignored-warning archetype: the VCs who screamed about Biden’s mild EO are now silent as Trump actually bans a model. The real story is the weaponization of export controls as a political cudgel, with Amazon acting as both investor and informant. The defensive nature of the jailbreak suggests the NSA or other agencies may have wanted to suppress a tool that helps defenders, not attackers. Expect more such bans as the administration learns it can make technology disappear with a single national security claim.
Date: June 15, 2026 02:06 PM ET
URL: https://www.techdirt.com/2026/06/15/the-vcs-who-screamed-that-biden-would-kill-powerful-ai-models-seem-quite-chill-about-trump-actually-doing-it/
AI Sentiment Score: Negative (85%)
AI Credibility Score: 10.0/10 — High
Scores and text generated by AI analysis of the source article indicated.The Community Repair Hub Is Live: Find a Repair Event Near You! (Ifixit)
Summary: iFixit, in partnership with U.S. PIRG and The Restart Project, has launched the Community Repair Hub, a centralized platform to find and list free repair events across the United States. The hub aggregates events from Repair Cafés and Fixit Clinics, offers an API for syndication, and provides free general liability insurance to qualifying nonprofit groups. This move addresses two persistent barriers to grassroots repair: discoverability of decentralized events and venue insurance requirements. The hub is live now, with groups like the Boulder U-Fix-It Clinic already posting full 2026 calendars.
Why it matters: This is a hackerspace-to-mainstream pipeline signal: iFixit is formalizing the infrastructure of the repair movement, turning ad-hoc volunteer events into a discoverable, insured network that can scale.
Context: Repair Cafés and Fixit Clinics have operated for years as decentralized, volunteer-run events, but lacked a unified calendar and often failed to secure venue insurance, limiting their growth and reach.
"For years, Repair Cafés and Fixit Clinics have quietly helped people fix what they already own. Volunteers gather in libraries, community centers, and maker spaces. Someone walks in with a broken lamp." — IFIXIT
Commentary: The insurance provision is the real signal here: iFixit is using its corporate liability policy as a backstop for the entire U.S. repair ecosystem, removing a structural bottleneck that no amount of volunteer enthusiasm could solve. The API availability also matters—it means the hub can feed into other calendars and tools, making repair events a first-class data layer rather than a walled garden. This is infrastructure, not just a directory.
Date: June 16, 2026 08:40 PM ET
URL: https://www.ifixit.com/News/117940/community-repair-hub-launch
AI Sentiment Score: Positive (66%)
AI Credibility Score: 10.0/10 — High
Scores and text generated by AI analysis of the source article indicated.The UK’s Teen Social Media Ban Is Political Theater, Not Child Safety Policy (Techdirt)
Summary: Discord would fall under the designation given for the law, but despite how popular and “controversial” it’s been for the past year or so, it’s odd that it’s not part of the “the ban will include” list. The UK’s Teen Social Media Ban Is Political Theater, Not Child Safety Policy from the techlash-on-demand dept UK Prime Minister Keir Starmer is not having a particularly good time of being the UK’s leader. Basically everyone thinks he’s doing a terrible job and it seems unlikely that he’ll be in the role much longer.
Why it matters: This matters for Hacker Community because it gives a concrete current signal to track: Discord would fall under the designation given for the law, but despite how popular and “controversial” it’s been for the past year or so, it’s odd that it’s not part of the “the ban will include” list.
Context: Discord would fall under the designation given for the law, but despite how popular and “controversial” it’s been for the past year or so, it’s odd that it’s not part of the “the ban will include” list. The UK’s Teen Social Media Ban Is Political Theater, Not Child Safety Policy from the techlash-on-demand dept UK Prime Minister Keir Starmer is not having a particularly good time of being the UK’s leader. Basically everyone thinks he’s doing a terrible job and it seems unlikely that he’ll be in the role much longer.
"Discord would fall under the designation given for the law, but despite how popular and “controversial” it’s been for the past year or so, it’s odd that it’s not part of the." — TECHDIRT
Commentary: The immediate implication is operational rather than speculative: watch how this changes budgets, workflows, or risk assumptions over the next cycle.
Date: June 16, 2026 04:14 PM ET
URL: https://www.techdirt.com/2026/06/16/the-uks-teen-social-media-ban-is-political-theater-not-child-safety-policy/
AI Sentiment Score: Negative (85%)
AI Credibility Score: 10.0/10 — High
Scores and text generated by AI analysis of the source article indicated.Tips on Writing a Good Manual (And How Not to Do It) (Ifixit)
Summary: The EU’s right-to-repair mandate has forced manufacturers to produce repair manuals, but quality varies wildly. iFixit’s guide contrasts excellent examples like Mackie’s engaging, concise manuals with failures like Teenage Engineering’s opaque PDFs and a fertilizer company that replaced printed instructions with a QR code leading to a 40-second YouTube video. The piece argues that video should supplement, not replace, text-based manuals, and that good manuals require technical understanding, clarity, and audience awareness.
Why it matters: As repair mandates expand, the quality of documentation becomes a de facto barrier to repair—bad manuals can nullify the right to repair in practice, making this a digital-rights fight over access to usable information.
Context: The EU’s right-to-repair legislation, effective since 2021 for certain appliances and expanding, requires manufacturers to provide repair manuals. iFixit has long advocated for repairability and published its own tech writing handbook for repair guides.
"Thanks to the EU, tech companies must now provide repair manuals for their products. But there’s a huge gap between creating a manual, and creating a good manual. Some companies (we’re looking." — IFIXIT
Commentary: This is a repair fight signal: the EU mandate creates a compliance checkbox, not a usability standard. The QR-code-as-manual pattern is a surveillance-resistance tool in reverse—it forces users into a data-hungry, non-searchable format. Expect hackerspaces to start publishing ‘manual audits’ that shame manufacturers into providing real PDFs, and for the community to treat QR-only documentation as a red flag equivalent to glued-in batteries.
Date: June 15, 2026 11:14 AM ET
URL: https://www.ifixit.com/News/117842/tips-on-writing-a-good-manual-and-how-not-to-do-it
AI Sentiment Score: Negative (75%)
AI Credibility Score: 10.0/10 — High
Scores and text generated by AI analysis of the source article indicated.Field Notes from a Year of OPSEC Training (Eff)
Summary: EFF published a detailed field report from a year of providing pro bono OPSEC training to at-risk communities and liberation movements. The piece documents a surge in demand from abortion access activists, trans rights advocates, sex worker rights organizers, and climate justice groups since the fall of Roe v. Wade. EFF distinguishes its approach from traditional pentesting by centering threat modeling, digital footprint mapping, and non-technical mitigations before introducing encryption or secure tools. The report identifies common threat actors—doxxers, law enforcement, platform censorship, insider infiltration—and argues that comprehensive data privacy legislation is an existential necessity for movement safety.
Why it matters: This is a rare, transparent account of how frontline digital security work actually operates outside the infosec vendor model, revealing the concrete threat landscape and resistance strategies used by the most targeted communities today.
Context: EFF has long provided OPSEC training but until now did not broadcast the details; this post is explicitly intended as a template for other trainers and organizers to copy.
"Since the fall of Roe v. Wade, we’ve seen a huge uptick in abortion access activists like clinic escorts and information distribution networks reaching out. So too are providers of criminalized healthcare services, both abortion services and gender affirming care alike. The list goes on: advocates for transgender rights such as art collectives and archivists, sex worker rights activists, survivors of intimate partner violence, climate justice activists, legal defense groups focusing on immigrant justice and Black liberation." — EFF
Commentary: The piece functions as an ignored-warning signal and a digital-rights fight archetype: it documents how surveillance infrastructure directly enables state and vigilante violence against marginalized groups, and how under-resourced movements are forced to rely on pro bono support rather than institutional security. The explicit call for other trainers to copy their methodology suggests a deliberate effort to scale community defense outside traditional infosec channels. The tension between publicity and privacy—movements needing visibility while avoiding doxxing—remains the central unsolved design problem for any organizing toolchain.
Date: June 18, 2026 03:00 PM ET
URL: https://www.eff.org/deeplinks/2026/06/field-notes-year-opsec-training
AI Sentiment Score: Negative (75%)
AI Credibility Score: 10.0/10 — High
Scores and text generated by AI analysis of the source article indicated.Flock Is The New Tool Of Choice For Cops Who Love Stalking Their Exes (Techdirt)
Summary: A new report from the Institute for Justice and follow-up reporting by 404 Media documents that police officers are systematically abusing Flock Safety’s automatic license plate readers (ALPRs) to stalk ex-partners. In one case, an Orange City, Florida officer ran his ex-girlfriend’s plate 69 times in a single summer, used the system to catch her in lies, and invited a colleague on a stakeout. The abuse is not being caught by internal police audits but by victims, journalists, and activists using public records requests and tools like HaveIBeenFlocked.com. Flock’s audit logs are proving useful for accountability, but only because the public is doing the work that departments refuse to do.
Why it matters: This is a concrete, repeated failure mode for surveillance infrastructure: giving cops ALPR access without meaningful oversight turns a public safety tool into a domestic abuse enabler, and the burden of detection falls entirely on victims and civil society.
Context: Flock Safety is the dominant ALPR vendor in the US, and its systems are now ubiquitous in suburban and small-town police departments. The pattern of officers using law enforcement databases to stalk exes is decades old; Flock just makes it easier, faster, and harder to hide.
"For months during the summer of 2024, Jarmarus Brown, an Orange City, Florida police officer, ran his ex-girlfriend’s license plate through the Flock automated license plate reader (ALPR) system lookup database at least 69 times. He searched for the license plate belonging to her mom at least 24 times, and searched for the license plate belonging to her dad at least 15 times." — TECHDIRT
Commentary: The key signal here is not just the abuse—it’s that the detection mechanism is entirely external. Police departments are not auditing themselves; victims are checking HaveIBeenFlocked.com and filing public records requests. This is a surveillance-resistance tool archetype: the same audit logs that enable stalking also enable accountability, but only if civil society has the resources to mine them. Expect more states to follow the Wisconsin model where victims can self-check their own plates, and expect Flock to face growing pressure to implement real-time abuse detection rather than post-hoc log dumps.
Date: June 15, 2026 12:27 PM ET
URL: https://www.techdirt.com/2026/06/15/flock-is-the-new-tool-of-choice-for-cops-who-love-stalking-their-exes/
AI Sentiment Score: Negative (80%)
AI Credibility Score: 10.0/10 — High
Scores and text generated by AI analysis of the source article indicated.Post ID: bd876f25