Tuesday, April 14, 2026

IMF Urges Nations to Stay at Frontier of Mounting AI Risks

During the IMF's Spring Meetings, Tobias Adrian, the Financial Counsellor, urged governments to remain vigilant against the escalating threats posed by artificial intelligence. He emphasized the necessity for proactive policy frameworks that enhance cybersecurity and ensure operational readiness to tackle potential risks, particularly in light of concerns surrounding the capabilities of Anthropic PBC's latest AI models.

Lagarde, Worried About AI, Lauds Anthropic’s Approach on Mythos

European Central Bank President Christine Lagarde has expressed strong support for Anthropic PBC's decision to limit the release of its latest artificial intelligence model, Mythos. In a recent interview, she underscored the critical need for enhanced safeguards around AI technologies, labeling the control of such advancements as a pressing concern for the economy and society at large.

Altman-Backed Startup Raises $170 Million for Robocar Networks

Glydways Inc., a robocar startup supported by Sam Altman, has successfully raised $170 million in a Series C funding round and is now negotiating for an additional $250 million. The company, which is eyeing a valuation of over $1 billion, is exploring more than 20 projects globally, particularly attracting interest from investors in Japan, including Obayashi Corp.

Financial risk management platform Pillar raises $20M seed in round led by a16z

Pillar, a financial risk management platform founded in 2023, has raised $20 million in a seed funding round led by Andreessen Horowitz. The company automates hedging processes for commodity-driven businesses, which have faced significant volatility in recent times due to geopolitical factors. Co-founder and CEO Harsha Ramesh emphasized the platform's goal of providing institutional-grade risk management tools to small and medium-sized enterprises, making hedging as accessible as payment systems.

Workers who want AI training the most are using it the least

A recent analysis by New York Federal Reserve researchers reveals that while workers who desire AI training are often those with lower incomes, they are using AI tools the least. The study highlights a stark disparity in AI adoption rates, with only 15.9% of workers earning less than $50,000 utilizing AI compared to 66.3% of those making over $200,000. This raises concerns about increasing inequities in the labor market as many workers express fears about job security in the face of AI advancements.

Italy court allows class action against Meta over Facebook data scraping

A Milan court has approved a class action lawsuit against Meta Platforms, stemming from a data scraping incident that compromised the personal information of approximately 533 million Facebook users globally between January 2018 and September 2019. The CTCU consumer association is representing affected users in Italy, where an estimated 35 million individuals may have lost control over their data, in violation of the EU's GDPR. Meta has expressed disagreement with the court's decision, emphasizing that it is merely procedural and does not imply any legal wrongdoing.

What a United-American merger would mean, from antitrust hurdles to airfare

United Airlines CEO Scott Kirby has suggested a merger with American Airlines, which would create the world's largest airline. However, this potential deal faces major antitrust hurdles, as the combined entity would control approximately 40% of domestic capacity, raising concerns among regulators about market concentration and fare increases.

CEOs are betting AI will augment work rather than displace all workers

At the Semafor World Economy conference, Anthropic co-founder Jack Clark expressed optimism about AI's role in the workforce, arguing that it will augment rather than replace jobs. He dismissed concerns that AI could lead to a 20% unemployment rate in the next five years, suggesting that the technology will fundamentally change business practices and societal interactions, necessitating a shift in employee skill sets.

Adobe fixes PDF zero-day security bug that hackers have exploited for months

Adobe has fixed a critical vulnerability in its Acrobat DC and Reader DC applications, which hackers have been exploiting for at least four months. This zero-day vulnerability, tracked as CVE-2026-34621, enables attackers to remotely install malware by tricking users into opening malicious PDF files. Adobe has urged users to update their software to protect against potential breaches, as the extent of the impact remains unclear.

Instacart acquires Instaleap to expand its enterprise platform internationally

Instacart has announced its acquisition of Colombia-based grocery technology firm Instaleap, a move aimed at expanding its enterprise platform internationally. Founded in 2019, Instaleap provides global fulfillment solutions, serving clients in nearly 30 countries. This acquisition allows Instacart to grow its international presence without the need to establish a delivery network in those regions.

Ukraine captures a Russian position using only drones and ground robots

In a significant development, Ukraine's military has captured a Russian position solely using drones and ground robots, marking the first such operation in the ongoing conflict without any infantry soldiers involved. President Volodymyr Zelenskyy announced this milestone, highlighting the use of seven different robotic systems that executed over 22,000 missions in the first quarter of 2026, resulting in the surrender of Russian forces without any casualties on the Ukrainian side.

Has Google’s AI watermarking system been reverse-engineered?

A developer known as Aloshdenny claims to have reverse-engineered Google DeepMind's SynthID watermarking system, suggesting that AI watermarks can be stripped from images or added to others. Google refutes these claims, stating that SynthID is a robust tool designed to prevent such manipulations. Aloshdenny's method involved generating 200 images and employing signal processing techniques, though he admits he could not fully remove the watermark, indicating its effectiveness.

Ecovacs’ new robovac spots and pretreats dried stains using powerful jets

Ecovacs has introduced the Deebot X12 OmniCyclone, an advanced robovac that utilizes AI to detect and treat dried stains on floors. This model enhances cleaning efficiency by spraying a mixture of water and cleaning solution before scrubbing, allowing it to tackle tough messes with fewer passes. The X12 is available for $1,499 and includes a sophisticated dock that automates maintenance tasks.

Anthropic Opposes the Extreme AI Liability Bill That OpenAI Backed

Anthropic has publicly opposed Illinois bill SB 3444, which OpenAI supports, as it would protect AI labs from liability for significant harm caused by their systems. This disagreement highlights a growing divide in AI regulation approaches between the two leading companies, with Anthropic advocating for accountability and safety measures. Despite the bill's low chances of passing, it has sparked intense lobbying efforts and discussions about the future of AI legislation in the U.S.

Claude Mythos is a wake-up call for Europe's AI safety apparatus

Anthropic's decision to restrict access to its AI model Claude Mythos has raised significant concerns about Europe's AI safety framework. With only a select group of 52 organizations, primarily US tech firms, granted access, European regulators are left with minimal insight into the model's capabilities. The UK's proactive testing of Mythos has already indicated its potential to enhance cybersecurity, underscoring the urgent need for European authorities to establish better access and oversight mechanisms for such powerful AI tools.

‘Google app for desktop’ launches on Windows

Google has officially launched the 'Google app for desktop' for Windows users worldwide, following a testing phase that began in September. This application aims to deliver the best of Google Search directly to users' desktops, featuring a keyboard shortcut for quick access to web searches, Google Drive, local files, and installed applications. The app also includes an AI Mode, allowing users to utilize Google Lens for specific searches and share their screens for contextual inquiries.

Coming soon: 10 Things That Matter in AI Right Now

In response to the overwhelming number of significant AI candidates, MIT Technology Review is launching a new annual list called '10 Things That Matter in AI Right Now.' This list will highlight not only cutting-edge AI technologies but also important trends and research directions in the field. The inaugural list will be unveiled on April 21, 2026, during the EmTech AI conference at MIT, with online publication following the event.

SAP brings agentic AI to human capital management

SAP has launched its 1H 2026 release of SuccessFactors, integrating agentic AI into human capital management to enhance operational efficiency and reduce costs. This update features AI agents that proactively identify and resolve administrative bottlenecks across recruiting, payroll, and talent development, streamlining processes and improving data accuracy while supporting compliance with regulatory standards.

Jim Cramer Says Market Has It All Wrong on Cybersecurity and AI

Jim Cramer contends that the market is mistakenly categorizing cybersecurity stocks like CrowdStrike, Palo Alto Networks, and Palantir alongside traditional software, ignoring how AI actually enhances their market potential by introducing new vulnerabilities. Despite their stock prices declining between 12-26% this year, these companies are experiencing significant growth in annual recurring revenue, indicating a disconnect between market perception and business performance.

A sleek, wearable airbag for cyclists is nearly here

Van Rysel is developing a revolutionary skinsuit with an integrated airbag system for cyclists, aiming for consumer availability within two years. This innovative design, created in collaboration with airbag technology specialist In&motion, is currently undergoing testing with professional riders and promises to deploy an airbag in just 60 milliseconds upon crash detection, significantly enhancing rider safety.

Samsung just raised prices on nearly a dozen Galaxy phones and tablets by up to $280

Samsung has implemented price increases on nearly a dozen Galaxy smartphones and tablets, with hikes reaching up to $280. This move follows a previous increase on the Galaxy Z Fold 7 and reflects the broader trend of rising costs in the tech industry, particularly due to shortages in storage and memory components.

Bringing people together at AI for the Economy Forum

At the AI for the Economy Forum in Washington D.C., Google and MIT FutureTech convened experts to explore how AI will shape jobs and the economy. The event emphasizes the need for collaboration among companies, workers, and governments to harness AI's benefits while mitigating risks. Google announced new investments in research and training programs aimed at equipping individuals with the necessary skills for a changing economic landscape.

Canada’s Scotiabank preps for its AI future

Scotiabank has introduced Scotia Intelligence, an AI framework designed to unify data and AI operations, enabling employees, particularly those in client-facing roles, to leverage AI under strict governance and security protocols. This initiative aims to enhance decision-making and operational efficiency, as evidenced by AI handling over 40% of client queries in contact centers and automating 90% of commercial email tasks.

ADT ‘My Safety’ turns your phone into an always-ready, portable security system [Gallery]

ADT has introduced 'My Safety,' a new feature for its ADT+ service that enables users to turn their smartphones into portable security systems. This feature includes tools like a Safety Timer for proactive monitoring and an Emergency Phrase that allows users to discreetly alert ADT's monitoring team by speaking a secret phrase. Available to all ADT+ subscribers, 'My Safety' enhances personal safety while leveraging ADT's existing resources.