June 1 Roundup: Google Launches Gemini Omni, Anthropic Overtakes OpenAI in Valuation, OpenAI Builds Robots, Illinois Mandates AI Audits, and Big Tech Kills Federal Oversight
Welcome to June. The AI industry arrived at the start of summer with moves that will echo for months: a new multimodal model that can edit video through conversation, a valuation flip at the very top of the AI hierarchy, a robotics push from the company that just shed everything that wasn't language, a landmark state law that puts frontier labs under independent audit, and a White House reversal that handed Big Tech exactly the regulatory landscape it wanted. Here is everything that mattered from the past 24 hours.
1. Google Introduces Gemini Omni — AI That Creates From Any Input
Google launched Gemini Omni on June 1, a new model family that fundamentally extends what AI can do with multimedia. Where previous models could understand images, audio, and video, Omni is Google's first model built to generate high-quality video grounded in real-world knowledge — and to let you edit that video through natural conversation.
The first model in the family, Gemini Omni Flash, is rolling out immediately to Google AI Plus, Pro, and Ultra subscribers globally through the Gemini app and Google Flow, and at no cost to YouTube Shorts and YouTube Create App users starting this week. The practical capabilities are significant: you can take a video you shot and ask Omni to change what's happening in it, add characters or objects, transform a moment into something unexpected — all through back-and-forth conversation where "every instruction builds on the last."
"With Omni, you can combine images, audio, video and text as input and generate high-quality videos grounded in Gemini's real-world knowledge. You can also easily edit your videos through conversation."
Google says the model maintains character consistency, physical coherence, and scene memory across edits — which has been the critical failure point for every prior video-generation system. Future Omni models will add image and audio output modalities. The announcement comes on the heels of Google I/O 2026, which unveiled Gemini 3.5 models, Project Aura smart glasses, and Google Search custom dashboards — marking one of the most aggressive AI product weeks in Google's history.
Video editing by conversation is the missing unlock for small and mid-size creative teams. The constraint on AI-generated video has never been the raw generation quality — it has been the inability to iterate without starting over. If Gemini Omni delivers on its "every instruction builds on the last" promise, it collapses the gap between "this is almost right" and "this is done." For marketing teams, agencies, and product teams creating video assets, this is worth testing immediately. Businesses that integrate Omni workflows in Q3 will have a meaningful cost and speed advantage over competitors still exporting raw shots to human editors for every revision cycle.
2. Anthropic Is Now Worth More Than OpenAI — Here's What That Actually Means
In a blog post published Thursday, Anthropic confirmed it "has raised $65 billion in Series H funding led by Altimeter Capital, Dragoneer, Greenoaks, and Sequoia Capital, valuing the company at $965 billion post-money." OpenAI's most recent valuation, set during its own funding round two months prior, stands at $852 billion. For the first time, the maker of Claude is nominally more valuable than the maker of ChatGPT.
The milestone is remarkable — and comes with significant asterisks. Anthropic claims to have just turned an operating profit for one quarter, as the Wall Street Journal reported, but that story also noted "it is unclear what accounting methods Anthropic has used to book revenue and costs," and that "the company might not remain profitable for the full year as it plans spending increases due to its vast computing needs." Those computing needs are enormous: Anthropic has committed hundreds of billions of dollars to Amazon, Google, and Broadcom over the next decade, and it is currently spending an estimated $1.5 billion per month on SpaceX compute capacity.
"Anthropic is now technically more valuable than OpenAI, 'the ChatGPT one.' There are, however, some mitigating factors." — Gizmodo
Investors appear to be pricing in explosive revenue growth rather than current profitability. Multiple analyst notes point to Anthropic's enterprise traction — particularly in regulated industries like legal, finance, and healthcare — as the driver of conviction at a near-trillion-dollar valuation. Meanwhile, CNBC reporting indicates that both OpenAI and Anthropic are on IPO paths, with Anthropic potentially going public as early as Q4 2026.
The valuation flip matters less as a scoreboard and more as a signal: enterprise trust in Claude has reached a level that large institutional investors are willing to price above OpenAI. That's a meaningful shift from 18 months ago, when Anthropic was viewed primarily as a safety-focused challenger. For businesses evaluating AI vendors, this round raises the question of strategic lock-in — both companies will now have IPO incentives to monetize customer relationships more aggressively. Enterprises signing multi-year AI agreements should ensure flexibility clauses are in place before either company's S-1 lands.
3. OpenAI Is Building Personal Robots — And Hiring Fast
CEO Sam Altman announced that OpenAI is actively hiring full-stack hardware, operations, systems, and machine learning engineers for a dedicated robotics division. The hiring push is the most concrete signal yet that OpenAI intends to compete in physical AI — not just as an investor in robotics companies, but as a builder.
The division grew out of OpenAI's world simulation research program, led by Aditya Ramesh, the researcher known for his work on DALL-E. Altman was explicit about the long-term vision: "In the short term, we are focused on robots to support skilled workers to build our future infrastructure; in the long term, we imagine everyone having a personal robot doing anything they need."
OpenAI has a complicated robotics history. In 2019, it launched a project featuring a robotic hand solving a Rubik's Cube, then shut the team down in 2020 to focus on large language models. Since then, it has invested in robotics companies including 1X, Figure, and Physical Intelligence — though its partnership with Figure to develop AI models for humanoid robots ended in early 2025. This internal rebuild is therefore OpenAI's third distinct approach to the physical world, and the first one with consumer ambitions at its center rather than research demonstrations. Morgan Stanley estimates the humanoid robotics market could exceed $5 trillion by 2050.
"In the short term, we are focused on robots to support skilled workers to build our future infrastructure; in the long term, we imagine everyone having a personal robot doing anything they need." — Sam Altman
The robotics announcement is part of a broader OpenAI hardware push that includes the $6.5 billion acquisition of Jony Ive's hardware startup io Products in May 2025, with consumer AI devices expected later in 2026.
OpenAI is betting that the company which controls the AI "brain" will eventually control the AI body. The near-term play — robots for skilled trades and infrastructure construction — is strategically smart because it targets labor markets where supply is already constrained and unions are less organized than in manufacturing. The personal robot framing is likely years away, but the hiring signal tells you something important: the AI race is entering its physical phase. Businesses in logistics, construction, and industrial operations should begin evaluating which workflows will be robot-addressable in a 3-5 year horizon, because OpenAI, Figure, 1X, and a dozen Chinese competitors are all converging on that timeline simultaneously.
4. Illinois Passes America's Strongest AI Safety Law — First Mandatory Third-Party Audits for Frontier Labs
The Illinois House of Representatives passed the Artificial Intelligence Safety Measures Act on May 27 by a unanimous 110-0 vote, sending the bill to Governor J.B. Pritzker, who is expected to sign it. When enacted, Illinois will become the first state in the nation to require independent, third-party audits of the safety practices of large frontier AI developers — a requirement that will apply directly to OpenAI, Anthropic, and Google DeepMind.
The bill, SB 315, mandates annual independent audits of frontier AI labs' safety protocols. Notably, both OpenAI and Anthropic publicly supported the legislation, a marked contrast from the industry's earlier resistance to state-level AI oversight. OpenAI spokesperson Jamie Radice said: "The Illinois General Assembly has shown real bipartisan leadership in advancing SB 315 and developing a thoughtful framework for frontier AI safety." Anthropic's head of state and local government relations, Cesar Fernandez, added: "Illinois is on track to become the first state to require independent, third-party audits of large frontier AI developers' safety practices."
"The Artificial Intelligence Safety Measures Act would be the first in the nation to require third-party audits of frontier model safety protocols." — Transparency Coalition
The Illinois bill lands in an increasingly fragmented regulatory landscape. Colorado recently passed SB 26-189, a narrower replacement focused on automated decision-making rather than frontier AI broadly. CNN filed a copyright and trademark lawsuit against Perplexity AI on May 28, alleging the AI search company unlawfully scraped and redistributed more than 17,000 of its news stories, photos, and videos. And OpenAI simultaneously published its Frontier Governance Framework, mapping internal safety practices onto the EU AI Act's Code of Practice — signaling that proactive compliance is now part of the company's go-to-market strategy in ways that lobbying against regulation was not.
The 110-0 vote in Illinois — bipartisan, unanimous — suggests AI safety audits have reached the same political untouchability as childhood safety standards. No legislator wants to be on record opposing them. The fact that OpenAI and Anthropic endorsed the bill rather than fought it tells you the frontier labs have decided that transparent audits, run on their own terms, are preferable to the alternative: state-by-state patchwork rules that could impose conflicting technical requirements across fifty jurisdictions. For enterprise buyers, third-party audits will eventually become a procurement checkbox the same way SOC 2 compliance is today. Start building those evaluation criteria into your AI vendor selection frameworks now.
5. Silicon Valley Lobbied Trump Directly — And Killed Federal Pre-Release AI Oversight
Hours before President Trump was set to sign an executive order on Thursday requiring government safety reviews of new AI models before their release, he abruptly reversed course. The Guardian, the Washington Post, Politico, and Axios all reported the same mechanism: tech billionaires including Elon Musk, Mark Zuckerberg, and former White House "AI czar" David Sacks made personal phone calls to Trump and persuaded him to pull the order.
"I didn't like certain aspects of it, I postponed it," Trump told reporters in the Oval Office. "We're leading China, we're leaving everybody, and I don't want to do anything that's gonna get in the way of that lead."
"The tech industry retains its ability to pursue rapid advancement of AI regardless of the potential harms, and Silicon Valley's leaders have successfully tested their power to kill any attempts at regulation in infancy." — The Guardian
The reversal is significant beyond this single executive order. It reveals the structural reality of AI governance in the current U.S. environment: the industry's ability to access the Oval Office directly means that any federal oversight mechanism must be pre-cleared by the same companies it would regulate. The White House had released a National Policy Framework for Artificial Intelligence on March 20, 2026, with a legislative blueprint designed to establish federal standards that would override state-level AI regulations — but that framework was explicitly industry-aligned. The killed executive order would have added a national security review dimension that even the framework didn't include.
Meanwhile, Jason Calacanis and other tech investors have argued publicly that genuine AI productivity gains are now driving workforce reductions — not "AI washing" — and urged affected workers to immediately adopt AI-first workflows. Peter Diamandis, in his latest Moonshots podcast, framed the same dynamic as an "organizational singularity" where AI agents, AI-native workflows, and recursive self-improvement are restructuring companies faster than workforce retraining programs can respond.
The killed executive order effectively confirms that the U.S. federal government will not impose pre-release oversight on frontier AI models in this administration. That means the regulatory action to watch is at the state level — Illinois, Colorado, California — and the international level, where the EU AI Act's compliance deadlines are still running. For multinational businesses, this creates an asymmetric compliance environment: you may face audit requirements in Europe and Illinois while facing none at the federal level. Build your AI governance stack around the strictest applicable jurisdiction, because the patchwork is not going to simplify in 2026.
6. Meta Launches Incognito AI Chat for WhatsApp — Privacy Push Amid Encryption Scrutiny
Meta introduced Incognito Chat for WhatsApp and the Meta AI app this week, offering temporary AI conversations processed in a secure environment that Meta claims even it cannot access. The feature is framed as a privacy-first mode for users who want AI assistance without their conversations stored or used for model training.
The timing is notable: the launch comes as Texas filed suit against Meta alleging that the company misled its 3.3 billion WhatsApp users about encryption protections, and as 29 states have filed briefs supporting the FTC's antitrust case against Meta. The company appears to be using Incognito Chat as a signal of privacy seriousness at precisely the moment its privacy practices are under maximum legal scrutiny.
"Meta says Incognito Chat is coming to WhatsApp and the Meta AI app, offering temporary AI conversations processed in a secure environment that even Meta cannot access." — Yahoo Tech
The feature also arrives as Meta faces intensifying pressure over its decision to remove Instagram's end-to-end encryption in certain contexts — a move that consumer advocates called a step backward for user privacy even as the company marketed Incognito Chat as a step forward.
Incognito AI Chat is a product designed as much for regulators and plaintiffs as it is for users. When you are simultaneously defending yourself against a state AG lawsuit over encryption and an FTC antitrust action, launching a "we can't even see your conversations" mode is a legal strategy as much as a user feature. For businesses using WhatsApp as a customer communication channel, the more important development is that Meta AI is deepening its integration into the messaging layer — which means your customers may be getting AI-assisted responses before they ever reach your team. That changes the expectations your support and sales workflows need to meet.
7. AI Beats the Average Human on Creativity Tests — And the Workforce Implications Are Becoming Undeniable
A massive new study comparing more than 100,000 people with today's most advanced AI systems delivered a finding that will land differently depending on your industry: generative AI can now beat the average human on certain creativity tests. The research, published through ScienceDaily, examined divergent thinking, associative fluency, and problem-reframing tasks — the cognitive functions most commonly described as "uniquely human" and therefore AI-resistant.
The timing aligns with a broader shift in how the technology industry is discussing AI and employment. Jason Calacanis stated publicly this week that genuine AI productivity gains — not "AI washing" — are now driving workforce reductions across the tech sector. He urged workers in affected roles to immediately adopt AI-first workflows as the only viable adaptation path. Peter Diamandis framed the same dynamic in his Moonshots podcast as the "organizational singularity": a threshold beyond which AI agents and recursive self-improvement restructure companies faster than any human workforce program can respond.
"Calacanis advises threatened workers to immediately adopt AI-first workflows." — Digg
The All-In Podcast community has simultaneously been observing that AI-native college graduates — those who grew up building with AI tools rather than learning to use them — are producing work that impresses experienced investors and operators in ways that feel qualitatively different from previous graduating classes.
When AI clears the average human bar on creativity tests, the strategic response is not despair — it is recalibration. The creativity floor has moved. That means work that was "good enough" at the human-average level now competes with AI outputs that are also good enough, but faster and cheaper. The value premium shifts to creativity that is genuinely distinctive, deeply domain-specific, or emotionally resonant in ways that require lived experience. For hiring, for talent development, and for content strategy, this is the year to audit which creative outputs in your organization fall in the AI-replaceable middle and which are genuinely above it. The ones in the middle need a plan.
Why This Week's Moves Matter
Five converging signals define the state of AI as June opens. First, multimodal generation has cleared the "useful in production" bar — Gemini Omni's conversational video editing is not a demo, it is a workflow tool, and it is free to YouTube creators starting this week. Second, the AI valuation race has reset: Anthropic now leads OpenAI nominally, and both are on IPO paths, which means enterprise AI contracts are about to become more expensive as investor return expectations harden. Third, OpenAI's robotics push marks the beginning of the physical AI era — the same models powering software agents will power physical agents, and the labor market disruption implications are multiplicative. Fourth, AI safety regulation is hardening at the state level even as it retreats federally — enterprises that build governance frameworks around the lowest common denominator will get caught when Illinois, Colorado, and California compliance requirements propagate to their vendor contracts. And fifth, AI is now measurably outperforming average humans on creativity tasks — which resets the baseline for what "skilled creative work" means in every industry. The organizations that thrive in the next 18 months will be the ones that take all five of these signals seriously rather than waiting for each to become unmistakable.
Need help navigating AI for your business?
Our team turns these developments into actionable strategy — from vendor selection and governance frameworks to hands-on implementation.
Contact SEN-X →