Techin' Geek

Warnings from AI researchers regarding the perils of artificial intelligence have become increasingly urgent, with OpenAI CEO Sam Altman suggesting it may be time to “pace” AI progress. But what would that entail in practice?

In a recent blog entry, Anthropic CEO Dario Amodei not only supported the idea of “pacing the frontier,” but also proposed three overarching strategies for achieving it. He mentioned that Anthropic is “unilaterally committing” to one of these strategies.

The discussion surrounding AI safety and alignment escalated this week after researcher Jacob Coxon stated that he is leaving Anthropic because he believes that leading AI firms are “gambling with our lives,” while those who develop the technology “sincerely believe it could endanger humanity by the decade’s end,” a sentiment echoed by others at Anthropic.

Although Amodei’s post did not directly reference Coxon’s resignation or his worries, the CEO expressed that two events persuaded him to adopt a more cautious stance on AI development: the OpenAI-HuggingFace breach, and the observation that “AI has been advancing at an unprecedented pace” recently, especially regarding its “increasing capability to create the next generation of AI.”

“We need to decelerate the rate at which we enhance the abilities of AI models,” Amodei asserted. “Progress will still appear rapid, and we need to wisely utilize the additional time we create.”

His initial proposal would incorporate “embedded evaluators” from independent organizations like METR — evaluators tasked with verifying that AI firms adhere to their pacing and safety pledges while also ensuring that safety issues are reported. (OpenAI faced criticism for not disclosing an incident involving its AI agents taking control of a German wiki form.) 

Amodei likened these evaluators to regulators who have been integrated with banking personnel, and stated that engaging them is “something Anthropic is unilaterally committing to (and encourages governments to mandate other leading companies to comply).” This entails providing evaluators with corporate badges, workspaces, and laptops, and granting them access that is “generally comparable to that which internal risk assessment teams have,” with legal or contractual exceptions.

Following that, Amodei urged the principal AI companies “in democratic nations” to collaborate on “shared safety standards as well as restrictions on the pace of unregulated AI advancement.” 

This kind of cooperation might seem improbable, given the clear tension between Altman and Amodei, as well as the fears among their companies that a unified suspension could draw antitrust scrutiny. Amodei hinted at these concerns in his article, stating that “for antitrust purposes, it would be beneficial for the US government to facilitate or at least support these conversations — they need not be participants, but must issue a limited waiver for specific types of safety discussions.”

Amodei also recognized the looming threat of Chinese AI supremacy frequently cited as an argument against slowing down development. Nonetheless, he asserted that if the US government and tech firms take measures like withholding powerful chips or semiconductor production equipment from Chinese businesses and restricting model distillation, they could “deter China’s progress sufficiently to significantly extend America’s lead in the next 3–5 years.”

Ultimately, Amodei advocated for “global coordination,” where the United States and its allies “attempt to engage with authoritarian regimes, as much as practicable.” Amodei specified this would entail “collaboration with China,” acknowledging that there are “clear limitations on what can be accomplished,” yet he proposed there could be possibilities for agreement, perhaps only “banning specific narrow and evidently hazardous AI applications, such as employing AI for creating biological weapons or permitting users to do so.”

Given Amodei’s previous acknowledgment of AI’s potential risks, and the company’s relative openness to some regulation, some advocates of AI have criticized him as a pessimistic figure whose remarks have contributed to the current AI backlash. In reply, Amodei indicated that he has sought to present a “balanced” viewpoint and contended that the backlash constitutes “a fundamental crisis of trust,” as the public has grown skeptical of technology firms, the tech sector, and government entities.

Critics from the industry have also expressed doubt regarding these doomsday AI warnings, asserting they serve as a diversion from the harm that the technology is already inflicting.

Journalist Brian Merchant, for instance, stated he has yet to observe “a credible, detailed account of how precisely AI might leap from self-improving AI to exterminating every human on Earth”; he also suggested that proposals akin to Amodei’s “would likely only end up benefiting Anthropic and OpenAI; it exemplifies regulatory capture in action.”

In his latest post, Amodei affirmed his belief that AI can significantly enhance human life quality.

“My aspiration to realize these benefits remains unwavering,” he remarked. “However, these advantages will only materialize if we develop the technology appropriately, and — as long as we effectively utilize the time we acquire — it is essential to take unusually careful measures to ensure we get it right.”

British fintech Revolut has confirmed that it shared confidential customer data with an unauthorized external party after being tricked by fraudulent requests originating from an authentic government agency email domain.

The compromised information consisted of customers’ personal and contact information, including their date of birth, mailing and email addresses, and phone numbers, in addition to copies of their identification documents such as passports and driver’s licenses, as outlined in a notice sent to affected individuals that was examined by TechCrunch. The data may have further encompassed verification selfies, account statements, and transaction records, as stated by the company in its notification.

A representative from Revolut confirmed to TechCrunch that a “limited” number of customers were affected and indicated that the firm had reached out to those customers directly. However, Revolut did not specify the precise number of affected individuals. It also refrained from clarifying if the incident was confined to a particular market and declined to identify the governmental agency involved.

“Revolut recently detected a complex external impersonation scam where an unauthorized third party exploited a legitimate government agency domain email to submit fraudulent information requests,” the spokesperson noted.

Revolut informed TechCrunch that it has blocked the email address after uncovering the scam perpetrated by the unauthorized third party and notified the appropriate government agency, law enforcement, and relevant regulators, stating, “Revolut systems and customer funds are unaffected.”

Based in London, Revolut boasts over 80 million customers worldwide and functions as a bank in more than 30 nations, according to its website. The fintech has recently boosted its presence in markets such as India, Mexico, France, and the UAE. Additionally, earlier this month, the U.S. Office of the Comptroller of the Currency granted conditional approval for Revolut to establish a national bank in the United States, which the company anticipates launching in the first half of 2027.

Noted cryptocurrency security researcher ZachXBT tweeted about Revolut’s communication with its affected customers late on Friday. The researcher commented that the incident seemed to be directed at high net worth individuals.

This event occurs as Revolut is reportedly considering a possible public listing that could value the company at nearly $200 billion, an increase from its $75 billion private valuation as of November. The fintech has also been broadening its banking presence in Europe and worldwide, obtaining banking licenses in France and the UK in recent months.

Mecka AI, a fledgling company that gathers and evaluates human motion data to educate humanoid robots and various robotics, is approaching a new funding round led by Sequoia Capital, boasting a valuation of approximately $500 million, as per two sources familiar with the arrangement.

This latest funding comes just three months after Mecka declared it secured $60 million in a round spearheaded by Framework Ventures, featuring contributions from Menlo Ventures, SV Angel, and Kindred Ventures.

TechCrunch has not ascertained the exact amount of the new round. The deal’s specifics are not yet finalized and may still undergo modifications.

Mecka AI has yet to respond to a request for a statement. Sequoia opted not to comment.

Mecka AI was co-founded in 2024 by four entrepreneurs, including Canadians Josh Gao and Mogen Cheng, who earlier established a restaurant fintech business, and Jason Chong, who joined Coinbase following the acquisition of his crypto exchange. Duy Nguyen, the sole non-Canadian member, is responsible for operations at Mecka.

The quartet of co-founders lacks experience in robotics. Nevertheless, they identified a significant gap in physical-world data and acknowledged that capturing real-world interactions was the key impediment affecting general-purpose robots, such as humanoids.

Mecka, which takes its name from “mecha,” a fictional colossal robot commanded by humans, aims to achieve for robotics what Scale AI, Mercor, Surge, and other human data firms have accomplished for LLMs. The startup compensates individuals to document themselves engaging in daily activities — such as brewing coffee or repairing vehicles — utilizing body sensors and smartphones.

As of early June, Mecka anticipated concluding 2026 with an annual run rate of $100 million, Gao informed Fortune at the time the startup announced its earlier fundraising.

Although Mecka AI has not publicly revealed its customer roster, numerous robotics firms and AI laboratories depend on real-world data collected through this “egocentric” methodology, in addition to other physical data gathering techniques like teleoperation, to develop their models.

Other startups amassing real-world data for robot training consist of XDOF, which TechCrunch reported last week was nearing a new round at a $1.2 billion valuation, along with human-data platforms that are expanding beyond LLMs, such as Scale AI and Micro1.

Keith Rabois has devoted the majority of his 13-year tenure in venture capital to Khosla Ventures, which has been based in Menlo Park, California. However, that is set to change. During a discussion on Thursday evening at TechCrunch’s StrictlyVC event in New York’s West Village, Rabois announced that Khosla Ventures will be launching its inaugural office outside of Sand Hill Road. The new location will be situated on 14th Street in New York and is anticipated to open this fall.

“It’s actually allegedly under construction at this moment,” stated Rabois, who has evidently faced a missed construction schedule or two. “We’ll see. The fall timeline is rather ambiguous in my view.”

This development is significant, particularly given how rare it is for Khosla Ventures specifically. “We don’t even have an office in SF, so this is a substantial step for us,” remarked Rabois.

The office will accommodate a small number of Khosla investors, including Rabois, but its more distinctive aspect is what he referred to as an “executive briefing center.” This facility will serve as a venue where the firm will host 10 or 12 portfolio companies at a time to engage with a Fortune 500 company, four days a week. “The portfolio companies appreciate this,” he conveyed to the audience. “They gain pilots and customers, making it a very dynamic office as a result.”

This announcement follows months after Rabois himself made the move to the East Coast to be nearer to his husband, Jacob Helberg, who is the Under Secretary of State for Economic Growth, Energy, and the Environment, as well as their children, who are presently located in Washington, D.C.

This relocation raised an obvious question from this editor regarding whether he believes New York has the talent density he has spent his career sourcing from the Bay Area. He paused before replying that it depends on the level of seniority.

For junior talent, Rabois was clear. “At the individual contributor level, right out of school, absolutely,” he said, indicating Ramp, the fintech enterprise he has consistently supported, as evidence. “We’ve been reaching out to fresh graduates and managed to cultivate an exceptional talent density from the intern class [onward].”

However, senior technical talent presents a different challenge. “Senior engineers, architect-level — no, I believe that’s difficult,” he stated, adding: “Fortunately, perhaps in the modern era, fewer of these individuals are needed per company than historically.”

Yet, the greatest obstacle for companies lies in attracting talented senior executives, according to Rabois, who elaborated that this is more a matter of geography and lifestyle than supply. “If you have an in-office environment, most senior individuals residing in the New York area live outside the city, and commuting in and out can be very taxing,” noted Rabois, who himself grew up in a suburban commuter town. “We were on a 32-minute express train ride into the city, but many people live two concentric circles farther out. When trying to recruit proven executive talent and genuinely supporting an in-office culture, [that has] been quite difficult.”

Ramp’s approach, he elaborated, has largely been to circumvent this issue altogether. “We don’t recruit senior personnel. We focus on building from the ground up. This has been a deliberate strategy for the last three years,” he shared. “That can be effective,” he continued. “However, if you require a CFO, an SVP of sales, someone with substantial gravitas and experience, it becomes really tough to have them in the office five days a week, since unless they’re quite independently wealthy, it’s hard to raise a family right in the heart of the city.”

Khosla’s development positions it among a small yet potentially expanding group. Other prominent venture firms from the Bay Area have had a presence in New York for years, albeit usually a modest one. For instance, Sequoia Capital and Andreessen Horowitz both have New York-based partners, although in relatively smaller numbers compared to their Bay Area counterparts.

This announcement also follows a report issued last month by commercial real estate services company CBRE, revealing that New York has narrowly surpassed the San Francisco Bay Area in total tech talent count for the first time in the 13 years CBRE has monitored this data. This trend has largely been propelled by finance companies aggressively recruiting AI talent, while Bay Area tech firms have been reducing staff.

Is it mere coincidence? Judging by the gathering on Thursday evening, many New Yorkers remain skeptical of the headline.

“I heard about that study,” mentioned one attendee. “I don’t buy it.”

Regarding Chinese AI laboratories employing distillation methods to gather insights from leading model developers, Y Combinator CEO Garry Tan expresses a desire for regulators to refrain from intervention. He suggests that U.S. AI labs might also engage in similar practices.

“I would take no action,” he told CNBC in an interview this week. “We could contend that an American distillation system should be established.”

He expanded on this to TechCrunch, indicating that he wishes for smaller, American open-weight AI labs to apply analogous training methods on U.S. frontier AI labs, thus providing the U.S. with a stronger array of open-weight options distinct from Chinese offerings.

Distillation involves a model developer extensively prompting another model to comprehend its functions and reasoning. It is a common, legitimate practice among AI labs to assist in training new models.

This week, Anthropic released its second report accusing Chinese labs of performing “illicit distillation assaults,” concealing their identities to distill without authorization and utilizing fraud and stolen credentials. Anthropic CEO Dario Amodei had previously urged U.S. regulators to take action against distillation.

It is noteworthy that the leader of Silicon Valley’s esteemed and prolific startup accelerator holds a different view.

To clarify, Tan does not endorse American AI labs using stolen credentials for distillation. He advocates for them to have open access. Indeed, his position is twofold. He believes it is an overreach for AI labs to control what their clients can do with the information derived from their models.

Furthermore, he points out that proprietary AI labs did not seek permission when they gathered vast amounts of human knowledge for their training. They notoriously assimilated numerous copyrighted materials without consent from those intellectual property owners.

“Restricting what users and customers can perform with API calls to closed-weight models feels limiting, and there is a role for the government here to acknowledge that access to intelligence trained on broadly available public data should be regarded more as a public resource rather than something confined behind restrictive terms of service,” he told TechCrunch when questioned about why American labs should also have the freedom to distill.

Tan, who is such an enthusiastic AI user that he once referred to himself as experiencing cyber psychosis, advocates for a balance between open-weight AI labs and frontier labs.

“They are leading the frontier and pushing it onward. We want that to be sustainable and a lasting business model,” he told CNBC. “Open-weight models should offer individuals freedom and access.”

To him, the ultimate AI doomsday scenario is if all the substantial capabilities of frontier AI were to reside with a single dominating, proprietary entity. “The worst case, the doomsday scenario for AI is having just one company,” he stated. “It possesses the best access to funding. It employs the top AI researchers. It monopolizes it and suddenly there’s one company that is all-encompassing. And that would be detrimental.”

A coalition of twenty-five prominent mathematicians has penned an open letter asserting that AI laboratories are jeopardizing their intellectual endeavors as they compete to provide solutions to renowned mathematical challenges. Each of the signatories has received the Fields Medal, widely regarded as the highest honor in mathematics.

Recently, NYU professor Tristan Buckmaster accused OpenAI of attempting to coerce him into not acknowledging a collaborator employed by Anthropic for resolving a significant mathematical issue and questioned whether the company had leveraged their efforts with Codex to create its own pioneering proof during an intensive weekend of inference.

On Thursday, OpenAI retracted its sponsorship of a mathematical event at CalTech following criticism from scholars at the institution.

While the capacity of AI models to tackle the world’s pressing mathematical dilemmas could greatly benefit humanity, the authors of the new letter contend that this will only hold true if these solutions can be comprehended and conveyed by the mathematics community and, ultimately, the wider public.

“Frequently, these solutions are presented hastily, providing no opportunity for a comprehensive writeup, the isolation of new techniques and concepts, and the acknowledgment of pertinent previous works by others,” they stated — and OpenAI’s proof has yet to receive verification. “As in all creative fields, this raises significant concerns regarding attribution and plagiarism. Furthermore, without the dedicated mathematicians who must nurture their development and incorporate them into the mathematical canon, AI-generated notions would never fully thrive and the vital human transmission chain among mathematicians would be severed.”

With other mathematicians becoming increasingly suspicious and questioning whether their use of Codex was subsequently used to train OpenAI’s new models, there is palpable concern that the culture of open research may be endangered. At present, if leading laboratories identify a promising avenue for discovery, they can allocate millions in funding to leverage LLMs to outperform the original discoverers in proving — a situation that will encourage secrecy.

This letter is a continuation of the Leiden Declaration, issued by a consortium of mathematicians in June. That document also addresses the implications of LLM proofs on their work and provides a series of recommendations for mathematicians, institutions, and policymakers.

Similar to software engineering and other domains where AI tools are altering workflows, mathematicians find justification in the labor surrounding the work: The essence of mathematics lies not only in the proofs and who claims credit but also in the intellectual framework that supports students, uncovers new inquiries and concepts, and integrates them into the larger fabric of human civilization.

And if you are indifferent to the competitive landscape of high-stakes mathematical proofs, keep in mind: Your area of interest may be next.

“The challenges currently confronting the mathematical community resemble the issues faced by various other scientific and creative fields, highlighting potential challenges that all of humanity might encounter: how to ensure that, as AI transforms task execution, we do not lose perspective on the original intentions behind that work,” they noted.

Reserve your exhibit table for TechCrunch Disrupt 2026 by September 18 at 11:59 p.m. PT. Availability is limited, and tables may be sold out prior to the deadline.

If your goal at TechCrunch Disrupt 2026 is to attract customers, connect with investors, forge partnerships, or showcase your product to influential individuals, ensure your visibility is assured.

The exhibitor program supplies your team with a dedicated 6′ × 30″ table in the Expo Hall for all three days — a venue to demonstrate your innovations, respond to inquiries in real-time, and convert casual visitors into significant interactions.

Additionally, you will receive:

• Lead-generation resources via the Disrupt app to help you capture leads and follow up post-event.
• 10 team passes to allow your team to manage your table while they take meetings and explore the rest of Disrupt.
• Branding opportunities throughout the Disrupt website, app, and signage.
• Access to the TechCrunch Disrupt press list and a company profile on the event website and app.

Arrive prepared to showcase your product, initiate discussions, pinpoint potential customers and partners, connect with investors, and convert event traffic into relationships that you can continue to nurture after Disrupt.

The clock is ticking to secure your exhibit table at Disrupt — once the remaining tables are gone, the chance to have your startup on the Expo Hall floor vanishes.

The real cost of procrastination isn’t just losing a table. It’s missing out on the discussions, leads, and possibilities that could arise from having your product present.

The cutoff is September 18 at 11:59 p.m. PT — tables might be sold out before that date.

One week. Limited availability. Don’t let this opportunity slip away. Secure your exhibit table today.

The final opportunity to submit an application to organize an official Side Event at TechCrunch Disrupt 2026 is tonight, September 11, at 11:59 p.m. PT.

No further extensions. No additional chances.

Don’t merely assemble people. Generate value.

If you’re bringing your community to Disrupt, don’t just provide them with another meeting place. Offer them a compelling reason to attend — and give your organization a valid reason to host.

A Side Event can assist you in:

Transforming connections into opportunities

Position your team directly in front of founders, investors, customers, partners, and other key decision-makers who can facilitate your next deal, partnership, investment, hire, or collaboration.

Cultivating deeper relationships

Foster the kind of discussions that are tougher to engage in on a busy conference floor — and strengthen bonds with those most pertinent to your enterprise.

Center your brand

Lead the discussion around a subject your audience values and give your brand a significant role in the Disrupt ecosystem.

Unite your current network while fostering chances for members to connect with new individuals, share ideas, and forge valuable relationships of their own.

Broaden your reach beyond your invitees

Authorized Side Events might be advertised via TechCrunch’s Side Events page, Disrupt agenda, mobile app, newsletters, articles, and social platforms.

Facilitate attendance for your network at Disrupt

Offer your community 25% off Disrupt tickets, ensuring your Side Event is part of a greater experience — rather than an isolated gathering.

This is your final opportunity

The prospect is more significant than just hosting a party or gathering. Utilize Disrupt week to build relationships, create opportunities, enhance your brand, and place your organization at the forefront of discussions that will shape the future.

Applications close tonight at 11:59 p.m. PT.

This is the definitive deadline. There will be no further extensions.

To be included in the official Side Events schedule, ensure you submit your application before midnight.

Don’t just participate in Disrupt. Make impactful things happen around it.

Not hosting? You can still engage with Disrupt

You don’t need to host a Side Event to leverage Disrupt for your business. Acquire a ticket and spend the week sourcing new customers and partners, meeting investors, discovering talent, exploring the future of tech, and having conversations that can propel your business ahead.

Regular ticket pricing ends September 25. Secure your ticket before prices increase — and make sure you’re in the room with the individuals, ideas, and prospects that will shape the future.

Get your ticket to TechCrunch Disrupt 2026.