Anthropic is moving to make AI-generated writing easier to identify, potentially giving publishers, schools, and other institutions a new way to establish whether text was produced with its Claude artificial intelligence models.

The AI company said Monday that new Claude models will embed an “imperceptible watermark” directly into AI-generated text. The marking is designed to have no effect on the meaning or readability of the content and will remain attached to text when it is copied and pasted. Anthropic also said the watermark may survive some forms of editing.

The technology marks a significant step in the growing effort by AI companies to establish the provenance of synthetic content. Rather than relying entirely on statistical AI detectors, which attempt to determine whether a passage appears machine-generated, watermarking creates a signal at the point where the content is produced.

Anthropic said the feature is part of its commitments to greater transparency under the European Union’s AI Act. Claude models launched on or after August 2 will support the marking from launch, while the company is working to extend the capability to older models.

The watermark will apply to Claude-generated content worldwide, including text produced when Claude is accessed through cloud providers. Anthropic also plans to give third parties tools that can detect the markings, potentially allowing publishers, universities, schools and other organizations to verify whether material was generated through Claude.

The development comes as the publishing industry confronts a difficult question: how can editors distinguish between legitimate use of AI as an assistive tool and cases in which AI-generated material is presented as entirely human-authored?

That question has already produced high-profile disputes.

Last month, a book agent withdrew support for the crime novel “Call Me, I’ll Hide the Body” following concerns that its author may have used AI. Fourteen publishers had reportedly bid for the book, and a publishing deal had been completed before the agent apologized. The author, Jerry Falade, denied using AI to write the novel.

Earlier this year, Hachette withdrew Mia Ballard’s horror novel “Shy Girl” following allegations that the work contained AI-generated writing. Ballard told The New York Times that she had not used AI to write the book and said a freelance editor had introduced AI-generated material without her direct knowledge.

Anthropic’s watermark could give publishers another layer of evidence in cases like these. A detectable Claude watermark could establish that Claude-generated material had been incorporated into a document, even when the text itself does not contain obvious signs of machine generation.

That distinction matters because conventional AI detection is inherently difficult. Modern language models can produce prose that closely resembles human writing, while editing can further obscure the characteristics that automated detectors attempt to identify.

Watermarking approaches the problem from a different direction. Instead of asking whether text looks like it was generated by AI, a detection system can look for a signal deliberately embedded by the model provider.

Google has already pursued a similar strategy. Google DeepMind expanded its SynthID system to AI-generated text in the Gemini app and web experience in 2024, embedding an imperceptible watermark by adjusting the probability of words selected during generation. The company says SynthID can identify AI-generated content while preserving the quality and meaning of the text.

Google has since expanded SynthID across text, images, audio and video and introduced a detector designed to identify content generated with its AI systems.

Anthropic’s move therefore points to an emerging industry standard in which AI companies take greater responsibility for identifying the material their models produce.

But watermarking will not make AI-generated writing impossible to disguise.

Anthropic acknowledges that extensive editing, paraphrasing, translation or combining Claude’s output with human-written material can make its watermark undetectable. That limitation means the technology is unlikely to provide a definitive answer in every disputed case.

There is also an important question of what exactly a watermark proves.

The presence of a Claude watermark could demonstrate that Claude was involved in producing some portion of a document, but it would not necessarily establish that Claude wrote the entire work. Someone could use the model for proofreading, translation, restructuring, or other limited assistance and still leave behind a detectable signal.

For publishers and educators, that distinction is expected to become increasingly important as policies around acceptable AI use become more sophisticated. A university may prohibit students from submitting AI-generated assignments while allowing AI-assisted proofreading. A publisher may permit an author to use AI for research or editing while requiring the prose itself to be written by the author. A watermark alone cannot determine whether such rules have been violated.

The more consequential development may therefore be the creation of a broader provenance system around AI-generated text.

If major AI laboratories consistently embed detectable signals in their outputs and make detection tools available to third parties, publishers and educational institutions could eventually have a more reliable mechanism for investigating disputed material. It could also make it harder for users to present entirely AI-generated work as exclusively their own.

At the same time, the technology raises questions about privacy, false accusations and the treatment of legitimate AI-assisted work. A detectable mark could be useful as evidence, but institutions will need to establish clear standards for interpreting that evidence rather than treating a watermark as automatic proof of misconduct.

The yen hovered near the psychologically important 160-per-dollar threshold on Tuesday as the impact of the rare U.S.-Japan currency intervention at the end of July continued to fade, reviving speculation that authorities could be forced to intervene again if the Japanese currency weakens further.

The yen was last at 159.20 per dollar, after falling 0.9% on Monday. The move has erased almost half of the currency’s gains following the intervention, when the yen strengthened to a three-month high of 155.20 after previously tumbling to a 40-year low of 163.99.

The rapid reversal reveals the difficulty facing Japanese authorities. Intervention can temporarily alter exchange-rate dynamics, but sustaining a stronger yen ultimately requires a shift in the underlying forces driving the currency, particularly the interest-rate gap between Japan and the United States.

Traders are therefore focused on whether the dollar can break through 160 yen. A sustained move above that level could increase political and market pressure on Tokyo to act again, particularly if the yen’s depreciation begins to push up import costs and inflation.

“This week coincides with Japan’s Obon holiday period, when reduced market participation tends to lower liquidity, potentially increasing the risk of sharp market moves during thin trading hours,” said Masayuki Nakajima, senior strategist for fixed income, currencies and commodities at Mizuho.

“In particular, if (the dollar) were to break decisively above the psychologically important 160 (yen) level, concerns about intervention could intensify further,” he said.

The market positioning has already changed sharply following the intervention. Speculators cut their net bearish yen positions by $8.865 billion in the week to August 4, leaving the net short position at $3.604 billion, according to U.S. regulatory data.

That was the largest weekly reduction in bearish yen bets in more than 12 years.

The decline suggests that traders have become more cautious about betting aggressively against the yen, given the possibility of another official response. But analysts expect short positions to rebuild if the fundamental case for a weaker yen remains intact.

The key variable remains monetary policy.

Japan’s interest rates remain substantially below U.S. rates, making the yen vulnerable to carry trades in which investors borrow in low-yielding currencies and invest in higher-yielding assets elsewhere. Unless expectations for Japanese monetary tightening strengthen or U.S. rates decline sufficiently to narrow the interest-rate differential, intervention alone may struggle to produce a lasting appreciation.

Japan’s authorities also face the challenge of intervening in a market that can quickly absorb official purchases. The July operation demonstrated that coordinated intervention can produce a sharp initial move, but the subsequent depreciation shows how quickly investors can return to the underlying trade when policy fundamentals remain unchanged.

Australian Dollar Supported By RBA Stance

The Australian dollar, meanwhile, remained near an eight-week high after the Reserve Bank of Australia left its cash rate unchanged at 4.35%, in line with expectations, while signaling that further tightening remains possible.

The currency was last at $0.7054, close to its strongest level since mid-June.

The RBA has already raised rates by 75 basis points since February as it attempts to contain persistent inflation, with higher energy costs adding to price pressures.

The decision to leave rates unchanged while retaining a tightening bias gives the Australian dollar support because it keeps the possibility of higher domestic yields alive. That could become particularly important if other major central banks move toward easier monetary policy.

U.S. Inflation Becomes The Next Major Test

The broader currency market remained subdued as investors waited for a series of U.S. economic releases that could determine the next direction for the dollar and global interest-rate expectations. Wednesday’s consumer price index is the main focus, followed by producer prices on Thursday and retail sales on Friday.

The inflation data could provide the clearest indication yet of how the Iran war and higher energy prices are feeding into the U.S. economy. A renewed increase in oil prices could complicate the Federal Reserve’s policy outlook by pushing inflation higher even as geopolitical disruption weighs on economic activity.

The dollar index was broadly unchanged at 99.84, while oil prices remained near one-week highs amid fading expectations of a U.S.-Iran agreement to end the conflict. That combination is creating a difficult environment for currency traders. Higher energy prices can support the dollar through inflation and safe-haven demand, but they can also increase pressure on the Federal Reserve if they feed into consumer prices and weaken growth.

The euro was little changed at $1.1537, while sterling stood at $1.3499.

China’s yuan remained near a three-and-a-half-year high against the dollar, with the offshore yuan at 6.7484 and the onshore yuan at 6.7468.

For the yen, however, 160 remains the immediate fault line. A decisive move beyond that level would put the effectiveness of the recent intervention back under scrutiny and force traders to assess whether Tokyo is prepared to spend more reserves to defend the currency.

The coming U.S. inflation data could make that calculation even more complicated. Analysts say a stronger-than-expected U.S. inflation reading could lift Treasury yields and the dollar, increasing pressure on the yen, while softer data could narrow the U.S.-Japan rate differential and give Tokyo some relief without requiring another intervention.