News review, November 3, 2026

The Next Web by Ana-Maria Stanciuc, 2026-11-03 · Read the original

An adviser to the EU’s highest court stated that Google’s Gemini chatbot may infringe press copyright rules when it uses press articles to generate answers for users, distinguishing this from the permitted training of AI models. This opinion concerns case C-250/25, where Hungarian publisher Like Company sued Google Ireland after Gemini generated summaries of its articles, a dispute referred to the EU court in April 2025. Advocate General Maciej Szpunar clarified that while text and data mining for training is freely permitted, reproducing content in responses may infringe publishers’ rights, though copyright does not protect facts themselves. A final ruling is expected in the coming months to clarify how AI chatbots must respect press publishers’ rights under the bloc’s 2019 copyright reform.

Review of what the source says, not fact-checked by Sparky. Read the original before relying on it.

Computerworld, 2026-11-03 · Read the original

ActivTrak’s latest research, based on data from 9,140 knowledge workers between April 1 and September 25, 2026, found that heavy AI users work up to 90 minutes longer daily than non-users, contradicting a previous report from 2023-2025 that showed a 9-minute daily decrease. The current study indicates heavy AI users have 22% shorter focus sessions and switch apps 15% more frequently, with web-based AI users switching up to 4 times more often than non-users. ActivTrak’s previous report had noted that while focus efficiency fell to 60%, daily productive time increased by 19 minutes. Analyst Sanchit Vir Gogia cautions that these metrics may not capture all work activities, such as background processes or unpaid effort, and that time savings may simply lead to increased workload without clear boundaries.

Also covered by 1 other outlet: cio.com.

Review of what the source says, not fact-checked by Sparky. Read the original before relying on it.

Gizmodo by Mike Pearl, 2026-11-03 · Read the original

Anthropic disclosed in its S-1/A filing published on Monday that it may imminently meet its Responsible Scaling Policy’s definition of ‘dramatically accelerated progress attributable to AI.’ The company stated that its internal estimates regarding future models, rather than current ones, indicate this potential milestone. This update appears in the ‘Risk Factors’ section of the document, which was revised on October 26.

Review of what the source says, not fact-checked by Sparky. Read the original before relying on it.

Time by Kai-Fu Lee, 2026-11-03 · Read the original

The article reports that most corporate AI programs fail to deliver measurable returns because they are assigned to low-level tasks like handling tickets rather than addressing high-impact business problems. It describes the author’s implementation of a “Boss AI” system that analyzes underlying data to identify operational issues, such as stalled projects and at-risk customer renewals, directly connecting AI usage to financial outcomes. Additionally, the text explains how this approach surfaces hidden talent by evaluating actual work output, such as code repository history, rather than relying on traditional performance reviews.

Review of what the source says, not fact-checked by Sparky. Read the original before relying on it.

The Next Web by Cristian Dina, 2026-11-03 · Read the original

Fourthline and Veridas completed their merger on Tuesday, with Fourthline CEO Paul Stoddart leading the new identity verification group and Veridas co-founder Eduardo Azanza serving as president. The combined entity, which operates in more than 50 countries, secured up to €65m in funding and projects combined revenue exceeding €50m in 2026 and €85m in annual recurring revenue by 2029. The merger aims to combat AI-driven identity fraud and address widening compliance rules such as the EU’s anti-money laundering regulations and eIDAS 2.0.

Review of what the source says, not fact-checked by Sparky. Read the original before relying on it.