News You Can Use

Edition 21 · 1st - 14th July 2025

News You Can Use

Deep Dives

Three stories worth sitting with

BCG - AI at Work 2025: Momentum Builds, but Gaps Remain

BCG - AI at Work 2025: Momentum Builds, but Gaps Remain

What
Based on a 10K+ employee global survey, BCG's AI at Work 2025 report reveals critical insights about how generative AI is being adopted, or rather under-adopted, at work. While leadership and technology roles report frequent weekly usage, only around 51% of frontline employees use AI regularly. Key barriers include lack of sufficient training (only 36% feel adequately prepared) and limited access to the right tools, which leads to unauthorised "shadow AI" use (over one-third reported).
So what
This highlights the importance of user training, access, and leadership alignment that is crucial for adoption of tools like AGPT and our tech stack. To achieve meaningful impact, our rollout strategy should include micro-training, coaching, and clearly defined workflows, not just rollout communications. Potential to use this as leverage for our AI workshops - indicating the need for engagement and training across the business.

Freshfields launches Legal Tech Master's for New Lawyers

Freshfields launches Legal Tech Master's for New Lawyers

What
Freshfields has announced a fully funded LLM (Masters) in Law & Technology co-delivered with King's College London for incoming trainee solicitors. The firm covers full tuition and offers a £20,000 stipend.
So what
This sets a clear benchmark: clients and competitors increasingly expect deeper tech understanding among legal teams. We should highlight skills development such as AI prompt literacy, tool training, and innovation fluency as core to our own team's strength.

"I'm losing all trust in the AI industry" - The Algorithmic Bridge

"I'm losing all trust in the AI industry" - The Algorithmic Bridge

What
The article expresses deep scepticism about the AI industry's trajectory, highlighting problems like excessive hype, economic instability, talent churn, and unclear technical roadmaps to AI breakthroughs. It compares the current phase of AI to a repetition of the dot-com bubble.
So what
This helps frame why responsible use, transparency, and grounded expectations matter, both internally and with clients. We should continue centering our messaging around practicality, oversight, and user feedback, not tech jargon or grand promises.

Worth Reading

Everything else worth a click

Wordsmith - Legal AI Engineer Hub

Wordsmith's new community and content hub for the emerging "legal AI engineer" role. Useful if you're building an in-house team on this boundary.

Grok 4 launch and benchmark

xAI launched Grok 4 with reasoning, tool use and a larger context window. Claims top benchmark scores on several reasoning tests, though data governance and safety posture continue to hold most enterprise legal teams back.

Anthropic - Legora Customer Story

Case study of Legora running on Claude. Useful datapoint on how legal AI vendors pick and switch between underlying foundation models.

Legaltech.com - LegalTechLab

New accelerator offering up to $250k plus mentorship for early-stage legal tech founders. Another signal that capital into the category is not slowing.

METR - AI Experienced Developer Study

Surprising RCT finding: 16 experienced developers using AI tools took 19% longer to complete issues while believing they had gone 20% faster. Big wake-up call on subjective vs measured productivity.

AI at Work 2025 - BCG (PDF)

[Internal AG resource] BCG's 10k+ employee survey on GenAI adoption. Only 51% of frontline employees use AI regularly, 36% feel adequately trained, and "shadow AI" is widespread.

VentureBeat - Labs Warn AI Interpretability Is Slipping

Rare joint warning from OpenAI, DeepMind and Anthropic researchers that chain-of-thought monitoring - the main way we understand what frontier models are "thinking" - may not survive the next generation. Real safety and assurance implications.