Amnesty International condemned the July 15 decision of Delhi's Lieutenant Governor to invoke Section 3(2) of India's National Security Act (NSA), which effectively grants police the power to hold individuals in administrative detention without procedural guarantees.

New research from Meta's Oversight Board found commercial large language models (LLMs) from Anthropic, DeepSeek, Google, Meta, OpenAI and xAI are far less likely to criticize repressive leaders and governments. The research calls it "free speech infringements by proxy." Apps and material produced by these generative AI tools could actively reproduce restrictions, including on political criticism. It's unclear why these AI models are doing this—underscoring the need for companies to be transparent and build "human rights due diligence" into the process, before it’s too late. China, Thailand, Cambodia and Saudi Arabia topped the list of countries with the highest refusal rates when LLMs were prompted to create materials criticizing governments. 

Amnesty International on July 21 called on authorities to immediately release dozens of people detained in Egypt over their alleged involvement with Gen Z002, a Discord-based group used to discuss political issues, criticize government policies, and host dialogue about bringing meaningful change to the country.

Mahmoud Khalil, a prominent Palestinian rights activist, filed a lawsuit July 14 in US federal court in Manhattan alleging a "coordinated" conspiracy between senior Trump administration officials and three private organizations to target him for deportation over his advocacy work. Khalil is a former Columbia University graduate student who became the public face of the Trump administration's crackdown on pro-Palestinian campus activism.

Mexican President Claudia Sheinbaum said July 9 that her government plans to seek criminal complaints in the United States over the deaths of Mexican citizens while in immigration custody or during anti-immigration operations.

At the UN's Global Dialogue on AI Governance in Geneva, Nobel Peace Prize laureate Maria Ressa warned that artificial intelligence poses the threat of an "information Armageddon." Scholars of critical theory perceive that online culture reproduces a "cycle of epistemic injustice" by fostering "organized stupidity" that fuels extremism and authoritarianism. And the same technology now implements the grim material application of this degraded intellectual climate, with a UN Human Rights Council study especially accusing Palantir of being part of an "economy of genocide" through its enabling of Israel's AI-enhanced targeting in Gaza. Meanwhile, the pollution generated by the data centers that power AI is now "almost incomprehensible." Despite all this, UNESCO in Geneva glibly called for a "safe and inclusive AI that benefits all," and an openly boosteristic "AI for Good" confab was held alongside the Global Dialogue. The "anti-tech rebellion" against screens in school classrooms is a significant sign of hope, but the end of reading and death of literacy are already upon us. Moves toward greater regulation of AI are most significant as a catalyst for the conversation that needs to be had: how to launch an effective movement for abolition. In Episode 336 of the CounterVortex podcast, Bill Weinberg continues to make the case.

UN experts on July 8 condemned the sentencing of Pakistani human rights defender Mahrang Baloch to two terms of life imprisonment by a court in Quetta, describing it as a "travesty" of justice.

Amnesty International on July 6 condemned the conviction in Türkiye of environmental activist Esra Işık. Amnesty's deputy regional director for Europe, Esther Major, spoke on the broader impacts of the conviction on the fight to defend the environment: