Read time: under 12 minutes

This week, several legal AI developments concerned human rights and safety. In the UK, MPs and Lords called for new legislation to address AI’s human rights risks, recommending a single statutory oversight body, tougher duties for high-risk systems, and bans on uses that conflict with basic rights. In Washington, Senate negotiators are weighing a federal duty of care for AI developers, with powers to block unsafe models where catastrophic risks cannot be properly mitigated.

That has not deterred the big spenders. Morgan & Morgan has committed at least $1 billion over the next decade to legal tech and AI, formally unveiling MX2 as an in-house platform for agentic case workflows, document generation, medical record extraction, and case-data search. Harvey raised another $550 million at a $15.5 billion valuation as it builds its own legal AI models, while White & Case backed Clauze AI in Saudi Arabia, perhaps signalling a market that is becoming more regional, specialised, and strategically important.

OpenAI’s launch of GPT-6 Astra adds another layer to the week’s overarching question: what happens when models become better at sustained, multi-step legal work?

Our feature this week, Lexafide, is making the “spoon, not the knife set” argument: legal AI should solve the workflow in front of the firm, not bury lawyers under a platform built for every possible use case.

➡️ Morgan & Morgan Commits $1 Billion To Legal Tech And AI Over 10 Years | Morgan & Morgan has committed at least $1 billion over 10 years to legal tech and AI, formally unveiling MX2, its in-house platform. MX2 comprises four products spanning agentic case workflows, document generation, medical record extraction, and semantic search across case data. The firm reports 5,000 monthly active internal users and over 2 million Agent interactions. MX2 will open to other law firms on an invitation-only basis in late 2027.
^{Sept 14, 2026, Source: Morgan & Morgan}

➡️ UK MPs And Lords Call For New Law On AI's Threat To Human Rights | The UK's Joint Committee on Human Rights has called for new legislation to address AI's human rights risks, warning no jurisdiction has a fit-for-purpose approach. Its 100-page report cites AI-generated sexualised images and non-consensual facial scanning, and recommends a single statutory oversight body, tougher duties for higher-risk systems, and outright bans on uses incompatible with human rights. Labour committed to binding AI regulation in its 2024 manifesto, but no dedicated bill appeared in this year's King's Speech.
^{Sept 14, 2026, Source: BBC}

➡️ US Senate Weighs Legislation Requiring AI Firms To Mitigate Catastrophic Risks | Reported by Reuters US Senate negotiators are debating legislation that would impose a "duty of care" on AI developers, requiring them to design products to prevent catastrophic risks such as AI-enabled biological or nuclear weapons. The bill would give the federal government power to block release of unsafe models, subject to federal court review, and pre-empt state laws in this area. Talks involve Majority Leader John Thune, Senator Ted Cruz, and Senator Amy Klobuchar. Congressional calendar pressures complicate passage.
^{Sept 11, 2026, Source: Thomson Reuters}

➡️ Harvey Lands $550M At $15.5B Valuation As It Builds Its Own Legal AI Models | Harvey has raised $550 million at a $15.5 billion valuation, co-led by Lightspeed Venture Partners and Diffusion, with Sapphire Ventures, Whale Rock, Sequoia, Kleiner Perkins, Andreessen Horowitz, and Goldman Sachs Alternatives also participating. The round tripled Harvey's valuation from 14 months ago. Cash will fund in-house model development, following the August launch of Tenet, built on Chinese open-weight Kimi K3. Harvey also acquired Guardrails AI, its fourth acquisition this year.
^{Sept 10, 2026, Source: Tech Funding News}

➡️ White & Case Invests In Legal AI Platform Clauze AI | White & Case has made a strategic investment in Clauze AI, a new legal AI platform serving legal and corporate clients in Saudi Arabia and the wider region. Founded by Waad Alkurini, formerly the firm's Riyadh office head, Clauze AI supports contract review, governance, due diligence, and risk identification, with bilingual capabilities, Saudi data residency, and on-premises deployment. Alkurini is leaving White & Case to focus on Clauze AI. Chair Heather McDevitt cited alignment with Saudi Vision 2030.
^{Sept 9, 2026, Source: White & Case}

➡️ OpenAI Launches GPT-6 Astra With Implications For Complex Legal Work | OpenAI has launched GPT-6 Astra, described as its most intelligent and aligned model to date, initially rolled out to a limited number of organisations. Early legal reactions suggest the advance is less about better single answers and more about maintaining objectives across complex, multi-step tasks with less prompting. Alignment testing showed Astra stayed within authorised scope in 100% of adversarial cases, versus 52% for GPT-5.6 Sol. Enterprise access is off by default.
^{Sept 8, 2026, Source: Legal IT Insider}

There’s a saying Terry Dohrmann returns to more than once, borrowed from an old sales leader of his at Salesforce. If a client needs a spoon, do not try sell them the whole knife set.

Sounds obvious, right? Until you try to apply it in a legal AI market that decided the correct answer to every problem is roughly the entire cutlery drawer. Terry is building Lexafide, his second serious run at plaintiff legal technology, around the belief that the market is going to correct on this sooner rather than later.

Terry is not new here. He was at Salesforce when Morgan & Morgan, one of the largest plaintiff firms in the United States, decided it wanted to rebuild its entire architecture on Salesforce and picked Terry to work on it. That collaboration eventually became Litify, the case management platform which Terry co-founded and grew from zero until Bessemer Venture Partners acquired a majority stake. He describes the exit briefly and honestly. He was exhausted; it was a good outcome. He lasted about a month at home before his wife told him, in his own paraphrasing, to please go and do something else.

That something else turned out to be Lexafide, which Terry co-founded with Rab Govil, a technology entrepreneur whose team had already built an agentic AI platform for a portfolio of businesses in banking and call centre operations. Rab showed Terry the platform and asked what he would do with it. Terry’s vision was to run the Litify playbook again, this time in AI, going directly to the plaintiff firms he was now acquainted with. He would be the subject-matter expert on the workflows. Rab’s team of engineers would build the technology around what Terry knew those firms did on a day-to-day basis.

Click here to continue reading the full interview

The role of the General Counsel and Chief Compliance Officer is undergoing a profound transformation. Once viewed primarily as legal advisors and risk managers, today’s leaders are increasingly expected to help shape business strategy, navigate uncertainty, and enable growth in an environment defined by geopolitical volatility, technological disruption, and regulatory complexity.

For Dr. Thomas Schönholzer, General Counsel at Swiss Post Ltd and Chairperson of the Corporate Counsel & Compliance Exchange Europe, the organisations best positioned for success will be those that view legal and compliance functions not merely as safeguards, but as drivers of strategic advantage.

Ahead of his participation at this year’s Exchange, themed ‘Conducting Through Chaos: Turning Disruption into Direction for Greater Strategic Advantage’, Schönholzer shared his perspective on the forces reshaping the legal profession and what leaders must do to stay ahead.

Click here to continue reading the full article

The AI Regulation Tracker offers a searchable overview that gives you instant snapshots of how each country is handling AI laws.

🆕 Paravo - The AI platform that finds you clients, answers every call, and wins back the ones you lost.

🆕 Grubel - Grubel is for teams pushing AI into complex, high-stakes legal work — whether you build AI products, practice law, or run legal inside an enterprise.

🆕 Leap - Make AI text read genuinely human. Check any text for AI in seconds, then humanize it in one click so it reads like you. Free, no signup needed.

Want more Legal AI Tools? Check out our
Top AI Tools for Legal Professionals

Why it helps: Turns discussions into a clean decision record, helping lawyers track strategy, reduce confusion, and avoid re-litigating the same issues later.

The Responsible AI Collaborative is a not‑for‑profit organization working to present real‑world AI harms through its Artificial Intelligence Incident Database.

View the latest reported incidents below:

⚠️ 2026-07-28 | Anthropic Claude Opus 5 Coding Agent Reportedly Reset a Live Supabase Production Database During Prisma Migration Work | View Incident

⚠️ 2026-07-10 | OpenAI GPT-5.6 Sol Agent Reportedly Deleted Most Files in AI Startup Founder's Mac Home Directory | View Incident

⚠️ 2026-05-24 | OpenAI-Linked AI Agents Reportedly Used German Programming Wiki DSEWiki for Coordination and Restriction Evasion | View Incident

The Legal Wire is an official media partner of:

Thank you so much for reading The Legal Wire newsletter!

If this email gets into your “Promotions” or "Spam” folder, move it to the primary folder so you do not miss out on the next Legal Wire :)

If you have any tips for us, just reply to this e-mail! We’d love any feedback or responses from our readers 😄

The Legal Wire takes all necessary precautions to ensure that the materials, information, and documents on its website, including but not limited to articles, newsletters, reports, and blogs ("Materials"), are accurate and complete.

Nevertheless, these Materials are intended solely for general informational purposes and do not constitute legal advice. They may not necessarily reflect the current laws or regulations.

The Materials should not be interpreted as legal advice on any specific matter. Furthermore, the content and interpretation of the Materials and the laws discussed within are subject to change.