Amazon Layoffs Continue: Team Behind Its Top AI Models Hit by Job Cuts

ibtimes.co.uk Jul 23, 2026

Amazon Layoffs Continue: Team Behind Its Top AI Models Hit by Job Cuts

Curated by Shubham · TechCapsules · Jul 23, 2026

AI-summarised brief · reviewed before publication

Amazon has reduced staff within its Artificial General Intelligence unit, which develops the company’s proprietary Nova foundation models. This latest round of cuts contributes to over 30,000 corporate positions eliminated since October. The company did not disclose the specific number of roles removed or identify which internal teams absorbed the reductions. Affected U.S. employees receive 90 days of pay, benefits, and severance eligibility. A spokesperson stated that while large model development remains a priority, tighter focus on customer needs necessitated these eliminations. This workforce reduction coincides with Amazon’s planned $200 billion capital expenditure for the year, primarily funding data centers and AI infrastructure. The cuts occur just days before the company reports second-quarter earnings. Executive leadership has previously indicated that generative AI adoption would drive efficiency gains and reduce the total corporate workforce. Senior departures from the AGI division have been consistent, including the exit of former head Rohit Prasad.

💡 Why It Matters

· The simultaneous surge in infrastructure spending and workforce reduction reveals a strategic pivot toward capital-intensive automation rather than human-led development.
· This approach suggests Amazon intends to leverage its own AI tools to permanently shrink its operational headcount while scaling technical capabilities.

AI Amazon Capsules

Read full article on ibtimes.co.uk

Court backs AI’s ‘copy-right’, OpenAI unveils Presence, Google’s Cyber-Shield — Weekly AI…

OpenAI unveiled Presence, an architecture integrating AI agents into corporate systems, currently piloted by BBVA, SoftBank, and IAG. The company also launched ChatGPT Health, a HIPAA-compliant platform allowing users to sync wearables and upload medical data for structured summaries. Additionally, details emerged regarding Project Camellia, a planned 3.2-gigawatt data center in Georgia. In a significant legal development, the Delhi High Court ruled that training large language models on published news content constitutes fair dealing under Indian copyright law. Justice Amit Bansal denied an interim injunction sought by Asian News International, stating ChatGPT outputs do not substantially reproduce copyrighted material. This decision sets a precedent for AI training rights in Asia. Meanwhile, Google enhanced its Gemini model with new cybersecurity capabilities, positioning it as a digital security tool against emerging threats.

💡 The Delhi ruling establishes a critical legal framework for AI training data in Asia, potentially influencing global copyright interpretations. OpenAI’s enterprise and health integrations demonstrate a strategic pivot toward regulated, high-stakes sectors requiring strict security and compliance standards.

koreaherald.com · Jul 26

KAIST, Nvidia deepen AI collaboration with joint research facilities

The Korea Advanced Institute of Science and Technology announced an expanded research partnership with US chip giant Nvidia, focusing on physical and agentic artificial intelligence. The collaboration includes a $300 million joint research program over five years, aimed at developing next-generation AI technologies tailored to the Korean language and domestic industries. KAIST will establish the Nvidia-KAIST Joint AI Research Lab at its Kim Jaechul Graduate School of AI. This initiative combines Nvidia’s full-stack AI technologies and Nemotron open models with KAIST’s research expertise. The partnership provides $50 million in AI computing resources annually and supports at least 10 researchers yearly, offering internships and potential full-time positions at Nvidia. The announcement follows recent high-level meetings between South Korean officials and Nvidia CEO Jensen Huang, including Huang’s June visit to Seoul where he pledged AI research facilities. This strategic alliance aims to secure core AI technologies for South Korea and strengthen its industrial competitiveness in the global market.

💡 This deal cements South Korea’s status as a critical hub for specialized AI development, moving beyond generic model training to industry-specific applications. By integrating local language nuances and industrial needs, the partnership directly addresses the gap between global tech giants and regional economic requirements.

firstpost.com · Jul 26

‘Money won’t matter in 2036’: Elon Musk says AI will reshape the…

Tesla and SpaceX CEO Elon Musk predicts that artificial intelligence will fundamentally reshape the global economy within the next decade, asserting that money will become obsolete by 2036. In an interview with The Economist editor-in-chief Zanny Minton Beddoes, Musk argued that if AI and robotics generate overwhelming abundance, the traditional value of currency diminishes. He suggested governments should directly issue financial support to citizens, dismissing inflation concerns by claiming increased output would prevent price rises. Musk further predicted deflation would replace inflation as the primary economic challenge. Beddoes questioned the societal transition, raising concerns about job losses, wealth inequality, and political instability. She asked if massive redistribution or universal basic income would be necessary. Musk maintained that traditional economic assumptions may no longer apply, advocating for direct treasury checks to support those displaced by automation in this new, AI-driven landscape.

💡 Musk’s proposal to fund citizens through direct treasury checks challenges foundational monetary policies, suggesting central banks may need to abandon inflation targeting in favor of managing deflationary pressure from automated abundance. This radical shift forces policymakers to confront the political volatility of mass unemployment before technological displacement becomes irreversible.

abcnews.com · Jul 26

Cheaper, open and intelligent: Chinese AI models gain ground, as they make…

Chinese artificial intelligence models are rapidly gaining traction in the United States, driven by their affordability and efficiency. San Francisco-based technology leaders, including Mozilla’s chief technology officer Raffi Krikorian, have switched to Chinese systems like Moonshot’s Kimi K3 and Z.ai’s GLM-5.2 for daily tasks, citing superior speed compared to expensive American alternatives. Major U.S. companies, such as cryptocurrency exchange Coinbase, are also adopting these models to reduce operational costs. This shift has frustrated American tech giants and prompted political backlash. The Trump administration accused Moonshot of using covert methods to build K3 based on Anthropic’s technology. U.S. politicians and companies allege illicit “distillation” of proprietary models, claims Beijing rejects as groundless. Despite American-led restrictions on advanced AI chip exports and warnings of further sanctions from Treasury Secretary Scott Bessent, Chinese AI continues to appeal to independent developers globally. The rapid adoption underscores a shifting competitive landscape in the global AI race, moving beyond earlier disruptions caused by startups like DeepSeek.

💡 The adoption of Chinese AI by major U.S. firms demonstrates that cost efficiency is overriding geopolitical tensions and security concerns in the commercial sector. This trend challenges the effectiveness of export controls by proving that accessible, high-performance alternatives can still capture significant market share despite technological restrictions.

abcnews.com · Jul 26

Cheaper, open and intelligent: Chinese AI models gain ground, as they make…

Chinese artificial intelligence models are rapidly gaining traction in the United States, driven by their affordability and increasing efficiency. Tech professionals, including Mozilla’s chief technology officer, have switched to systems like Moonshot’s Kimi K3 and Z.ai’s GLM-5.2 for daily tasks, citing superior speed compared to expensive American alternatives. Major U.S. companies, such as Coinbase, are also adopting these models to reduce operational costs. This shift has frustrated American tech giants and prompted political backlash. The Trump administration accused Moonshot of using covert methods to build K3 based on Anthropic’s technology, while U.S. politicians allege illicit model distillation. Beijing rejects these claims as groundless. Despite U.S. restrictions on advanced AI chip exports and warnings of further sanctions to protect intellectual property, Chinese AI continues to appeal to independent developers globally, signaling a significant evolution in the competitive landscape between the two nations’ technology sectors.

💡 The adoption of Chinese AI by U.S. firms demonstrates that cost-efficiency can override geopolitical tensions and export controls. This trend challenges the effectiveness of American technology sanctions as a tool for maintaining dominance in the global AI race.

indianexpress.com · Jul 26

Silicon Valley splits over closing the borders to Chinese AI

A deep philosophical divide over artificial intelligence software development has intensified within Silicon Valley, reaching a boiling point this week. Leading AI firms Anthropic and OpenAI argue that models are too dangerous for open development and require tight corporate control for safety. Conversely, tech giants Microsoft and Nvidia contend that open-source AI is essential for innovation and business growth. Nvidia CEO Jensen Huang and Microsoft CEO Satya Nadella publicly endorsed open-source models, signing a letter alongside executives from Meta, Palantir, and IBM. Simultaneously, OpenAI and Anthropic have lobbied Washington regulators against Chinese open-source AI models, citing improper data harvesting. This conflict involves Treasury Secretary Scott Bessent and science adviser Michael Kratsios, who view American AI as valuable intellectual property. The dispute stems from rapid progress by Chinese startups Z.ai and Moonshot AI, whose models now rival American competitors, sparking intense debate over security, openness, and global competition in the AI sector.

💡 The clash reveals a fundamental tension between proprietary safety controls and the collaborative nature of technological advancement. As Chinese firms close the performance gap, the outcome will determine whether global AI development remains an open ecosystem or fragments into restricted, competing silos.