Week in review #44
Anthropic reports 45x coding productivity gain with AI.
Anthropic’s 45-fold coding productivity gain
In a preliminary IPO prospectus, Anthropic stated that its AI model, Claude, was responsible for at least 90% of the company’s own software coding in the preceding month. This has resulted in a more than 45-fold increase in the average amount of code written per engineer per day compared to the company’s 2024 levels. The filing demonstrates a significant, quantifiable productivity leap in a core technical function, showing how the role of software engineers is shifting towards prompting, reviewing, and testing AI-generated output.
aitoolsrecap.comfnnews.comindiatimes.com
HSBC and DNB cut jobs due to AI
HSBC is reportedly planning widespread job cuts in its UK wealth management business, attributing the move to increased automation from AI. The reductions are expected to affect a significant portion of its financial advisers, managers, and specialists. Separately, Norway’s largest bank, DNB, announced it is cutting 400 jobs, or nearly 4% of its staff. The bank stated that AI agents are already providing “substantial efficiency gains” in areas like customer data management, know-your-customer processes, and software development.
UK business AI adoption doubles in one year
A new report from the UK Department for Science, Innovation and Technology (DSIT) reveals that 32% of UK businesses were using at least one AI technology in 2026, a figure that has doubled from 16% in 2025. Among these adopters, 71% reported improvements in workforce productivity, and 60% noted more efficient processes. The growth is concentrated in accessible tools based on Large Language Models, with the most common applications being administration (81%), marketing (71%), and IT functions (66%).
Key EU AI Act rules now apply
Several key provisions of the European Union’s AI Act have come into force. Transparency duties for certain AI systems have applied since August 2026, and rules for general-purpose AI models have been effective since August 2025. A new ban on AI systems that generate non-consensual intimate material or child sexual abuse material (CSAM) will take effect on December 2, 2026. Deadlines for compliance with high-risk AI obligations have been postponed, with maximum fines for violations eventually reaching €35 million or 7% of worldwide turnover.
ai.gov.gralgorithmwatch.orgalicelabs.ai
Google releases Gemini 4 and partners with Foxconn
Google unveiled its latest foundation model, Gemini 4 Carbon, on October 28. The release continues the rapid pace of development among major AI labs. In a related move to accelerate infrastructure deployment, Google also deepened its strategic partnership with Foxconn. The collaboration aims to cut the time required to bring new chip supply into operational data centers from the typical 12–18 months down to approximately six months, addressing a critical bottleneck for the industry.
blog.googlefelloai.commashable.com
Mistral raises €3B, Microsoft adjusts spending
French startup Mistral AI has secured €3 billion in a Series D funding round, reaching a post-money valuation of over €21 billion. The round, led by Samsung Electronics, the EQT-managed Scaleup Europe Fund, and PSG Equity, is the largest ever for a private European technology company. Meanwhile, Microsoft adjusted its capital expenditure forecast for 2026 downward from $175 billion to $170 billion, citing delays in third-party service contracts. The company is also offering voluntary buyouts to 60,000 employees as it manages the high cost of its AI investments.