OpenAI Kills Sora, GPT-5.4 Launches, and AI Funding Hits Record Highs

OpenAI Kills Sora, Pivots to Robotics as Video AI Burns $15M Daily

In what may be the most dramatic product reversal in AI history, OpenAI announced this week it is shutting down Sora, its standalone video generation app and API, just six months after a splashy launch that briefly topped the App Store. The reason? Sora was burning through an estimated $15 million per day in inference costs while generating a mere $2.1 million in total lifetime revenue from in-app purchases.

The shutdown sent shockwaves through the creative industry. OpenAI stated that the Sora research team will refocus on world simulation research to advance robotics, viewing the underlying technology as more valuable for physical-world problem solving than media generation. For professional services firms that had begun exploring AI video for client presentations, marketing, and training content, this is a wake-up call: even the most well-funded AI products can disappear overnight.

Making matters worse, Disney ended its planned $1 billion investment in OpenAI, a deal announced just three months prior in December 2025. The collapse signals that even the largest enterprise partnerships are fragile when AI economics do not add up.

GPT-5.4 Thinking Arrives with Native Computer Use

While Sora exits the stage, OpenAI's new flagship model is making waves. GPT-5.4 Thinking, released on March 5, combines improved reasoning, coding, and agentic workflows into a single frontier model. It supports up to one million tokens of context and introduces native computer-use capabilities, enabling agents to operate computers and carry out complex workflows across applications.

Key improvements for professional services include 33% fewer factual errors compared to GPT-5.2 and significantly better token efficiency. The model can now provide an upfront plan of its thinking process, allowing users to adjust course mid-response. For law firms running document review, consulting firms building financial models, or medical practices analyzing patient data, this represents a meaningful leap in reliability.

GPT-5.4 also powers new built-in tools for spreadsheets, presentations, and documents, moving closer to the kind of end-to-end workflow automation that professional services firms have been waiting for.

Visa Launches Agentic Commerce: AI Agents That Buy on Your Behalf

The payments infrastructure for AI agents is no longer theoretical. Visa launched its Agentic-Ready Program this month, working with over 100 partners worldwide to enable AI agents to initiate and complete transactions on behalf of users. Banco Santander and Visa completed Latin America's first end-to-end payments powered by AI agents across five markets.

The implications for professional services are profound. Imagine AI agents that can autonomously procure office supplies, book travel, manage subscriptions, and handle routine vendor payments within predefined rules and budgets. Visa's Trusted Agent Protocol, introduced with over 10 partners, provides the security framework to distinguish legitimate AI agents from malicious bots during checkout.

For finance and consulting firms managing client accounts, this opens new advisory opportunities around agentic commerce governance: helping clients establish policies, authentication protocols, and compliance frameworks for AI-initiated spending.

AI Funding Shatters Records: $1.7 Billion in Two Weeks

March 2026 has already produced more $100M+ AI funding rounds than any comparable period in venture history. The numbers are staggering:

AMI Labs, founded by Meta's former chief AI scientist Yann LeCun, raised $1.03 billion in the largest European seed round ever, backed by Jeff Bezos, NVIDIA, Samsung, and Temasek. The company is building world models based on JEPA architecture, a fundamentally different approach to AI that could reshape how machines understand physical environments.

NVIDIA made a $2 billion strategic investment in Nebius for 5 gigawatts of AI factory capacity by 2030, confirming that AI infrastructure remains the safest bet in the ecosystem. Meanwhile, robotics companies including Mind Robotics ($500M), Rhoda AI ($450M), and Sunday ($165M) collectively raised over $1.2 billion in a single week for AI-powered industrial, household, and logistics automation.

On the enterprise side, Gumloop raised $50 million (Series B, led by Benchmark) for a platform that lets non-technical employees build AI agents for workplace automation, a direct threat to traditional consulting engagements around process optimization.

Legal AI Adoption Doubles as Governance Gaps Widen

A landmark report from 8am released this month reveals that nearly 70% of legal professionals now use generative AI for work, more than double the figure from just one year ago. Legal-specific AI adoption jumped from 21% to 42%, with firms reporting time savings of 6-20% per week and revenue gains of up to 20% attributed directly to AI tools.

However, the governance story is concerning. 43% of firms have no formal AI policy and no plans to create one, while only 9% have a policy that is actively enforced. Data security (46%), ethical issues (42%), and privilege concerns (39%) remain the top worries, yet most firms are not translating those concerns into formal safeguards.

The structural implications are also becoming clear: 41% of respondents expect new AI-specialist roles to be created, 39% predict a reduction in paralegal positions, and 21% anticipate fewer junior associate roles. For law firm leadership, the message is unmistakable: AI adoption is no longer optional, but deploying it without governance frameworks is a liability waiting to happen.

Professional Services Leads All Sectors in AI Implementation

According to McKinsey's latest research, professional services leads all sectors in generative AI adoption, with implementation rates soaring from 33% in 2023 to 71% in 2024. By 2026, the industry is moving decisively from pilots to production, fundamentally reshaping service delivery models.

Gartner predicts that through 2026, 20% of organizations will use AI to flatten their organizational structure, eliminating more than half of current middle management positions. Meanwhile, PwC's Global AI Jobs Barometer found that workers with AI skills command wage premiums up to 56% higher than their peers, even as job numbers rise in highly automatable roles.

The regulatory landscape is also hardening. States including California, Colorado, Illinois, and Texas have adopted new requirements governing AI in employment decisions. For consulting, legal, and HR advisory firms, this patchwork of state-level regulation represents both a compliance challenge and a significant new revenue opportunity in helping clients navigate the evolving rules.

What Professional Services Firms Should Do This Week

This week's developments point to several immediate action items for firm leaders across law, finance, consulting, and medical practices:

Reassess AI vendor dependencies. The Sora shutdown proves that even products from the most well-capitalized AI companies can be terminated with little warning. Diversify your AI tool stack and avoid building critical workflows around a single provider.

Evaluate GPT-5.4 for production workflows. The 33% reduction in factual errors and native computer-use capabilities make this model a strong candidate for document review, financial analysis, and research tasks that previously required human verification at every step.

Develop an agentic commerce policy. As Visa and others build the infrastructure for AI agents that make purchases, your firm and your clients need clear policies on authentication, spending limits, and audit trails before autonomous transactions become commonplace.

Close the AI governance gap. With 43% of law firms lacking any formal AI policy, the firms that establish clear guidelines now will have a competitive advantage in client trust and risk management.

Invest in AI-skilled talent. The 56% wage premium for AI skills is not a bubble; it reflects genuine productivity gains. Firms that train their existing workforce and recruit AI-literate professionals will outperform those that treat AI as a technology department concern.