Anthropic taps Google chip veteran Amir Salek as part of push into hardware

business-standard.com Aug 22, 2026

Anthropic taps Google chip veteran Amir Salek as part of push into hardware

Curated by Shubham · TechCapsules · Aug 22, 2026

AI-summarised brief · reviewed before publication

Anthropic PBC has appointed Amir Salek, a former leader of Google’s custom chip program, to its compute team as the AI company advances its strategy to develop in-house semiconductors. Salek, who oversaw the first seven generations of Google’s tensor processing units until 2022, will report directly to James Bradbury. This hiring marks a significant step in Anthropic’s effort to build an internal silicon business, moving beyond its current reliance on external suppliers like Nvidia, Google, and Amazon. The San Francisco-based firm, known for its Claude platform, has recently begun recruiting for this initiative. Like other major AI players, Anthropic faces intense pressure to secure sufficient data center infrastructure to support its rapid growth. Developing proprietary chips aims to mitigate supply chain shortages and allow for hardware designs specifically optimized for the company’s unique computational requirements, ensuring greater control over its technological foundation.

💡 Why It Matters

· Securing specialized silicon expertise allows Anthropic to bypass reliance on third-party suppliers, directly addressing critical bottlenecks in AI infrastructure.
· This move shifts the competitive landscape by enabling tailored hardware solutions that optimize performance for specific models.

Read full article on business-standard.com

Claude will now watermark all content generated using its tools

Anthropic has announced that all content produced with its Claude AI tools will now include a visible watermark. The change follows the company’s recent signing of the European Union’s Code of Practice on Transparency of AI‑Generated Content, which aims to clarify the origins of machine‑created material. The watermark will appear on text, images, and other media generated by Claude, ensuring that users can identify AI‑originated content. The move reflects growing regulatory pressure and industry efforts to address transparency and security concerns as AI output becomes harder to distinguish from human work.

💡 By embedding watermarks, Anthropic sets a practical standard for AI transparency that could influence how other firms label machine‑generated media, potentially reshaping trust and accountability in digital content.

spiegel.de · Aug 25

Amodei vs. Altman: How the Race for AI Dominance Is Increasing the…

Dario Amodei and Sam Altman, former colleagues at OpenAI, now lead rival companies Anthropic and OpenAI in the intense competition for artificial superintelligence. Their professional relationship shifted from collaboration to fierce rivalry after Amodei departed OpenAI to establish Anthropic as a major competitor. This dynamic was visibly highlighted during an AI summit in New Delhi, India, on February 19, where Indian Prime Minister Narendra Modi hosted both executives. Modi invited the tech leaders onto the stage for a group photograph, aiming to project unity and demonstrate progress in leveraging AI as a global growth engine. Despite the appearance of cooperation, the underlying tension between the two leaders underscores the high-stakes nature of the race. Amodei, once Altman’s head of research and a key engineer, now heads the organization challenging OpenAI’s market dominance. The summit served as a public backdrop for this private conflict, illustrating how personal and professional histories influence the broader trajectory of artificial intelligence development and corporate strategy in the technology sector.

💡 The personal history between these leaders transforms corporate competition into a high-stakes race where safety protocols may be sacrificed for speed. Their rivalry dictates the pace of superintelligence development, prioritizing market dominance over collaborative risk mitigation.

memeburn.com · Aug 24

Anthropic Expects Its IPO to Match or Beat SpaceX’s $86 Billion Record

Anthropic is preparing for an initial public offering that aims to match or exceed SpaceX’s $86.2 billion record, potentially setting a new benchmark for US IPO volume in 2026. The AI developer’s annualized revenue run rate surged to $65 billion by late July, up from $9 billion at the end of 2025. Managed by Morgan Stanley, Goldman Sachs, and JPMorgan Chase, the filing could occur by month’s end. Simultaneously, Anthropic is reversing its controversial data retention policy to allow enterprise customers to store data on their own cloud infrastructure. Despite reporting positive adjusted operating income in the second quarter, the company posted a net loss of nearly $42 billion in 2025 due to high compute costs. CEO Dario Amodei warned that growth misses could threaten survival. While some backers project a $2 trillion valuation, users on social media question whether the product justifies the high valuation amid concerns over rate limits and pricing.

💡 The simultaneous pivot to customer-controlled data storage directly addresses enterprise security concerns that previously hindered adoption. This strategic correction validates the massive capital raise by aligning product trust with financial expectations.

taipeitimes.com · Aug 24

Anthropic market debut could break SpaceX IPO record

Artificial intelligence startup Anthropic PBC is preparing for an initial public offering that could surpass the record-breaking SpaceX debut. Bloomberg reports the company expects to match or exceed SpaceX’s US$85.7 billion raise, potentially seeking over US$100 billion. This valuation would place Anthropic at US$2 trillion, more than doubling its previous US$965 billion worth. The New York Times indicates shares might list in autumn, allowing Anthropic to precede rival OpenAI to market. Founded in 2021, Anthropic positions itself as a safety-focused AI alternative. Its Claude Code product drives projected annual revenue of US$47 billion. To address computing power shortages, Anthropic hired Amir Salek, former head of Google’s custom chip program. Salek will lead efforts to develop in-house semiconductors, aiming to mitigate supply constraints and tailor hardware to specific needs. The company declined to comment on these developments.

💡 Anthropic’s potential US$2 trillion valuation signals a shift in market dominance from hardware pioneers to software-driven AI entities. By hiring ex-Google chip architect Amir Salek, the company is vertically integrating to secure its supply chain against global semiconductor shortages.

indianexpress.com · Aug 23

How developers are trying to remove Anthropic’s AI text watermarks

Anthropic announced plans to embed invisible, machine-readable watermarks in text generated by future Claude models to comply with the EU AI Act, which took effect on August 2, 2026. Within days of this announcement, developers released tools designed to strip these hidden markers from AI-generated content. One such tool by Guillaume Meyer gained significant traction on GitHub, attracting over 100 contributors. These removal utilities highlight a growing cat-and-mouse dynamic between AI transparency measures and evasion techniques. Anthropic acknowledges its watermarks only indicate likely involvement, not authorship, and cannot distinguish between fully generated and heavily edited text. The company states the markings do not affect content quality. However, widespread adoption of such detection methods raises concerns about false positives, potentially leading to unfair rejections of job candidates or unwarranted accusations against students and researchers regarding AI usage in their work.

💡 The rapid emergence of removal tools exposes the fragility of technical compliance solutions, suggesting that regulatory mandates may inadvertently fuel an arms race rather than ensuring transparency. This dynamic threatens to undermine trust in AI detection systems, creating legal and ethical risks for institutions relying on flawed verification methods.

completeaitraining.com · Aug 23

Anthropic lists AI backlash as risk factor in IPO filing ahead of…

Anthropic, the creator of the Claude AI model, is listing public concern regarding artificial intelligence and data center construction as a risk factor in its initial public offering prospectus. The company, valued near $1 trillion in private markets, is preparing for what could be the largest public offering on record, with investors anticipating a potential $2 trillion valuation. This disclosure follows a confidential filing in June and preliminary meetings with bankers in San Francisco. The risk factor reflects shifting public opinion, evidenced by a Gallup survey showing seven in ten Americans oppose local AI data center construction. Political pushback is intensifying ahead of midterm elections, with leaders in Florida and Pennsylvania implementing restrictions. Investors are questioning how competition, margin pressure, and potential infrastructure slowdowns might impact Anthropic’s growth. Compute capacity expansion remains critical for the company, which recently surpassed a $65 billion annual revenue run rate. The IPO filing provides insight into how leadership navigates these social and political headwinds that could affect infrastructure expansion and enterprise AI availability.

💡 The filing exposes how political resistance to infrastructure directly threatens the capital-intensive growth model of leading AI firms. It signals that regulatory and social friction, not just technological competition, now dictates the financial viability of the sector's largest players.