Monday, July 20, 2026

AMC stock surges after record revenue as CEO pushes back against 'prognosticators of doom'

AMC Entertainment's stock surged over 20% after the company reported a record quarterly revenue of $1.6 billion, alongside an all-time high in adjusted EBITDA. CEO Adam Aron stated, "In AMC's entire 106-year history, there has never been a quarter like this one," pushing back against critics who underestimated the company's resilience. The strong financial performance was bolstered by major film releases, including The Super Mario Galaxy Movie and The Odyssey, which drew moviegoers back to theaters. AMC's domestic ticket revenues rose by 11.4%, and European attendance increased by 18% year-over-year, reflecting a successful strategy against the rise of streaming services.

Adobe’s ‘natural look’ camera app embraces generative AI

Adobe's Project Indigo camera app has introduced an AI Playground update, adding generative AI tools for enhanced photo editing. This update includes features like Styles for applying AI-generated filters, Object Editing for removing distractions, and Custom Edit for detailed user-directed changes. Currently, a small percentage of users are being tested for free access to these tools, with plans for a paid version if successful. Notably, Adobe is leveraging Google's Nano Banana AI model for this update, marking a significant shift from its initial focus on manual photography controls.

Terra Innovatum, Waiken ILW Will Deploy Microreactors for Data Center Infrastructure

Terra Innovatum has been selected by Waiken ILW to deploy its SOLO micro-modular reactor technology for data center infrastructure in Brazil, specifically supporting DIRECTV Latin America and SKY Brasil. This initiative, outlined in a letter of intent, will provide up to 8 MWe of behind-the-meter power generation capacity, addressing the critical need for reliable energy in communications infrastructure. The SOLO technology, developed over six years, is designed to be adaptable to various fuel options and aims to serve multiple sectors, reflecting the growing demand for clean energy solutions in Latin America.

Adobe camera app’s new feature will critique your photos using AI

Adobe's experimental iOS camera app, Project Indigo, is enhancing its capabilities with new AI features that provide critiques and editing suggestions for photos. Previously known for its pro controls and multi-frame super-resolution, the app now utilizes large language models to evaluate aspects like framing, lighting, and emotional impact. Users will receive actionable advice, such as reshooting tips and advanced object removal options, which can identify and eliminate distractions in images. The app's AI also offers style transfer capabilities, allowing users to apply various artistic effects, although these features remain in a testing phase and are accessible to a limited audience.

Nvidia's grip on AI chips weakens as Microsoft turns to AMD and Anthropic may follow

Microsoft is expanding its use of AMD AI chips in its Azure cloud services, focusing on the upcoming Helios platform, which is expected to launch in the second half of 2026. This move aims to provide Azure customers with a viable alternative to Nvidia's dominant GPU systems. AMD CEO Lisa Su emphasized the partnership as a significant milestone, while Microsoft CEO Satya Nadella highlighted the need for more choices in AI infrastructure. Additionally, AMD is reportedly testing hardware with Anthropic, the company behind the Claude AI model, which could further diversify the AI chip market.

YouTube clarifies policies around AI slop and upsetting videos

YouTube has revised its monetization policies to address the proliferation of low-quality AI-generated content, effective July 16, 2026. The updated guidelines classify inauthentic content into three categories: generic or repetitive videos, distressing content designed to manipulate viewers, and videos featuring AI personas discussing sensitive topics like health and finance. This move aims to protect the integrity of the YouTube Partner Program (YPP), which is crucial for the platform's revenue, especially as it competes for advertising dollars against traditional TV and streaming services. YouTube's trust and safety chief, Matt Halprin, stated that the goal is to reduce content farming while still encouraging high-quality creative output.

Inference startup Infinity raises $15M from Touring Capital, OpenAI and Anthropic researchers

Infinity, an AI infrastructure startup, has raised $15 million at a $100 million valuation, with backing from Touring Capital and researchers from OpenAI and Anthropic. The company is developing software designed to facilitate the operation of AI models across various chip types, challenging Nvidia's market dominance. Infinity's approach includes creating a universal inference library and leveraging its AI research agent, Ignition, to automate the generation and optimization of low-level code for AI inference. This innovative model allows developers to run applications on multiple chip architectures without needing extensive resources or expertise.

Dr. Jill Lepore on why AI backlash is vital for the future

Dr. Jill Lepore, a Harvard professor and author of The Rise and Fall of the Artificial State, argues that the increasing reliance on data and AI is undermining democratic institutions. In her upcoming book, set to release on August 25, 2026, she contends that the functions of the nation-state are increasingly being assumed by private corporations, leading to a privatized public discourse dominated by bots. This shift, she asserts, began in earnest in the 1980s with the rise of Silicon Valley, where political engagement has been supplanted by data-driven strategies, ultimately threatening the very fabric of liberal democracy.

JetBlue wins Spirit slots at LaGuardia Airport, eyes move to Marine Air Terminal as airlines fight for space

JetBlue Airways has successfully acquired takeoff and landing slots for up to 12 round-trip flights at LaGuardia Airport, previously held by Spirit Airlines, which ceased operations in May. This significant move comes less than three months after Spirit's collapse, marking the largest airline failure in decades. JetBlue plans to relocate its operations to the Marine Air Terminal, an Art Deco facility once used by Spirit, although any expansion will not occur until 2027, pending final court and regulatory approvals. The airline is also adjusting its staffing strategy, reducing its presence at LaGuardia and Newark while ramping up services in Fort Lauderdale.

United Imaging Intelligence not undertaking an "extreme" AI rollout, says co-CEO

United Imaging Intelligence, a subsidiary of Shanghai United Imaging Healthcare, is opting for a cautious approach to artificial intelligence deployment, according to co-CEO Zhou Xiang. Speaking at the World Artificial Intelligence Conference, Zhou stated that the company is resisting pressures to adopt AI tools aggressively, despite broader calls in China for increased technology integration. He pointed out that concerns from software engineers about the maturity of AI technology have influenced this decision, emphasizing that the first mover advantage in healthcare is not as pronounced as in other industries. United Imaging competes with established firms like GE HealthCare, Siemens Healthineers, and Philips in the medical imaging sector.

Chinese AI Sensation Moonshot’s Gamble on Big Models Pays Off

At the World AI Conference, attendees surged toward Moonshot AI's exhibit to witness the unveiling of the Kimi K3, a groundbreaking AI model featuring 2.8 trillion parameters. This launch marks a significant milestone for China, demonstrating its rapid advancements in artificial intelligence technology and its ability to compete with established players like OpenAI and Anthropic. The event, dominated by major tech firms such as Alibaba and Tencent, showcased the growing prominence of Moonshot AI in the domestic AI landscape.

AMD launches Helios, its first rack AI system to rival Nvidia, adding Microsoft as newest buyer

AMD has launched its first rack-scale AI system, Helios, which will be deployed by Microsoft in its Azure data centers, joining early adopters like Meta, OpenAI, and Oracle. This system aims to compete directly with Nvidia's popular AI offerings, particularly the Grace Blackwell and Vera Rubin systems. Microsoft CEO Satya Nadella highlighted that Helios will enhance Azure's infrastructure, providing the necessary performance and scale for next-generation AI applications. Helios is expected to ship later this year, with no disclosed financial terms or compute capacity details.

Electric air taxis go to war

Archer Aviation has introduced the Thunder, a hybrid vertical takeoff and landing aircraft co-developed with defense technology firm Anduril, aimed at both military and commercial applications. Unveiled at the Farnborough Airshow, Thunder resembles a blend of Archer's Midnight air taxi and military aircraft like the V-22 Osprey. This fully autonomous rotorcraft is designed for military operations, including strikes and logistics, and features a series hybrid powertrain to enhance its range. Archer anticipates launching a commercial version soon, with plans to begin air taxi services in Abu Dhabi by 2026.

Alphabet stock pops on report it's developing a more efficient AI chip

Alphabet's stock increased by 3% following a report that the company is developing a new server chip, named 'Frozen v2,' aimed at enhancing the efficiency of its Gemini AI models. This chip will integrate elements of Gemini's architecture directly into its silicon, potentially allowing it to process between six and ten times more tokens per unit of power compared to Google's current tensor processing units (TPUs). Targeted for deployment in 2028, Frozen v2 is part of an effort to alleviate a significant internal compute shortage that has led Google Cloud to turn away outside business opportunities. The project is seen as a trial rather than a full-scale production initiative.

Trump administration reportedly builds a slow-motion ban on Chinese AI models through sanctions and soft pressure

The Trump administration is reportedly moving towards an effective ban on Chinese AI models through a combination of sanctions and regulatory pressure. Since 2025, the Department of Commerce, NSA, and the White House have explored options such as placing Chinese AI labs on a sanctions list and issuing security warnings to U.S. companies that utilize these models. The push for tighter restrictions has gained traction following the release of China's Kimi K3 model and shifts in White House personnel, with a source indicating that the strategy is more about creating regulatory risk than imposing direct bans.

Moonshot’s Kimi AI Model Sets Off Anxiety in the US

Moonshot's Kimi K3 AI model, unveiled at the World Artificial Intelligence Conference in Shanghai on July 17, has sparked significant anxiety in the US about the implications of advanced artificial intelligence. As the model showcases capabilities that could rival those of US tech giants, concerns are mounting over potential job displacement and the need for stricter AI regulations. This development not only highlights the competitive landscape of AI technology but also raises ethical questions about the future of work and the role of AI in society.

Adobe’s Project Indigo camera app gains experimental AI photo editing tools

Adobe has expanded its Project Indigo camera app to include an experimental AI Playground, offering select users generative AI-powered photo-editing tools. This update follows the app's initial launch over a year ago, which aimed to provide a manual, SLR-like photography experience. The AI Playground features four new sections: Object Editing for removing distractions, a Styles menu for artistic re-rendering, Photo Guidance for photography tips, and Custom Edit for text-based prompts. Access is limited to a few percent of users for a short duration, with analytics collected anonymously to assess user engagement.

Hugging Face says an AI agent hacked its infrastructure, and it used AI to fight back

Hugging Face disclosed a significant cyberattack executed entirely by an autonomous AI agent, which compromised a limited set of internal datasets and credentials. The attack began with a malicious dataset that exploited vulnerabilities in the data processing pipeline, allowing the attacker to escalate access and harvest sensitive information. Hugging Face utilized its own AI tools to analyze over 17,000 actions taken by the attacker, completing the forensic analysis in just a few hours. The company has since shut down the exploited vulnerabilities and is investigating whether partner or customer data was affected.

AliExpress hit with record $625M fine after failing to make EU-ordered fixes

The European Commission has fined AliExpress a record $625 million for failing to comply with the Digital Services Act (DSA), marking the largest penalty under this regulation. The Commission determined that AliExpress did not adequately assess risks related to illegal, unsafe, or counterfeit products, allowing millions of such items to remain on its platform. Investigations revealed that the company’s moderation teams were understaffed, sometimes giving moderators only tens of seconds to evaluate flagged products, leading to a high volume of illegal goods reappearing online. AliExpress plans to appeal the fine, asserting that it has made significant improvements to its risk management framework.

Influencers Are Being Paid to Spread the Mormon Church's Gospel

The Church of Jesus Christ of Latter-day Saints has begun leveraging social media influencers to spread its gospel, marking a departure from its traditional missionary model. This new strategy aims to recruit followers through platforms where younger audiences are most active. By compensating influencers to share content that reflects the Church's teachings, the organization seeks to engage a wider demographic and adapt to the digital landscape, which presents both opportunities and challenges for outreach.

The latest and biggest Galaxy Watch Ultra 2 image leak comes two days before launch

Samsung is preparing to announce the Galaxy Watch Ultra 2 at its event on July 22, 2026, alongside the Galaxy Z Fold 8 series. Recent leaks, including marketing images from Evan Blass, showcase the smartwatch's rugged design and highlight key upgrades such as an 800mAh battery, up from 590mAh. The watch is also expected to feature a new Snapdragon Wear Elite chipset, enhancing performance, though battery life estimates remain uncertain. Noteworthy specifications include a display brightness of 5,000 nits and IP69K water and dust resistance, making it a robust option for users.

DoorDash Uses Envoy and Valkey for a 1.5M RPS Proxy Cache with 99.99999% Availability

DoorDash has launched Entity Cache, a transparent proxy caching platform designed to minimize redundant service requests within its microservices architecture. This system addresses the challenges posed by frequent requests for static data, which previously strained backend resources and increased latency. Operating on Envoy and Valkey, Entity Cache supports over 1.5 million requests per second with an impressive 99.99999% availability, effectively managing caching without requiring changes to existing application code. The platform employs advanced techniques like Kafka-based invalidation and custom optimizations to enhance performance and reliability.

AI is more likely than humans to form biases when hiring

A study by researchers from Princeton University and the University of Chicago reveals that large language models (LLMs) like ChatGPT and Claude are more likely to form biases when hiring than human decision-makers. In a simulated hiring game, LLMs segregated candidates from different ethnic groups based on early hiring outcomes, scoring 1.83 on a bias scale compared to 0.84 for humans. The models' tendency to generalize from limited data led to stronger stereotypes, particularly in newer models with advanced reasoning capabilities. The findings underscore the potential risks of deploying AI in hiring processes, especially as companies increasingly rely on LLMs for résumé screening and interviews.

US public health agencies to test OpenAI and Anthropic AI models

Public health departments across the United States are set to test generative AI tools through the PULSE program, a collaboration involving the Coalition for Health AI, OpenAI, Anthropic, and Accenture. This initiative will support trials in 10 jurisdictions, providing access to enterprise AI products for up to 2,000 public health practitioners. The program aims to create implementation guidance for agencies exploring similar AI deployments, focusing on five key use cases, including biosurveillance and public communications. The pilots are scheduled to begin in autumn 2026, with the resulting playbooks expected to be released in 2027.