AI Reveals Hidden Identity in Renaissance Master’s Tudor Sketches

Artificial intelligence has possibly solved a long-standing mystery surrounding two Renaissance sketches by the master portraitist Hans Holbein, indicating that the identification of one portrait may have been incorrectly assigned some 300 years ago. Researchers at the Bradford University have analysed works from the Royal Collection and discovered that an unnamed woman depicted in one sketch could in fact be Anne Boleyn, Henry VIII’s tragic second wife, whilst the portrait traditionally thought to depict Boleyn may instead depict her mother, Elizabeth Howard. The discovery, made using cutting-edge artificial intelligence that analyses artistic characteristics across Holbein’s extensive body of work, questions assumptions that have persisted since the 1700s, when the sketches are believed to have been mislabelled.

The Mystery of False Attribution

The confusion surrounding these two Holbein sketches originates chiefly in inconsistent labelling practices that plagued the artist’s extensive corpus of work. Independent scholar Karen Davies, who commenced the investigation, uncovered notable variations among more than 80 Holbein works housed in the Royal Collection. Her initial investigation, published in March, revealed that under 15 per cent of the sketches have contemporary documentary verification—meaning most attributions depend on later inscriptions rather than original sources. This shortage of systematic historical verification enabled misidentification, particularly when labels were applied long after the works were created.

The Windsor sketch itself presented a puzzling contradiction to conventional identification. The portrait shows a sitter with light skin and auburn locks, characteristics that appeared at odds with documented accounts of Anne Boleyn, who was regularly recorded as having a deeper skin tone. Yet the sketch had been recognised as Boleyn’s portrait for centuries, primarily since no one had systematically questioned the 18th-century labelling. Davies observed that the Holbein corpus was notorious for such errors—one well-known instance concerned a portrait of Henry Howard that was incorrectly identified, when it depicted his father rather. These recurring mistakes suggested a wider issue with how the collection had been organised and preserved.

• Less than 15 per cent of Holbein works have documented contemporary evidence.
• The Windsor sketch’s physical characteristics contradicted conventional depictions of Anne Boleyn.
• 18th-century inscriptions produced misattributions lasting three centuries unchallenged.
• The Holbein corpus contained recorded instances of confused identities of family members.

How Artificial Intelligence Resolved an Artwork historical Mystery

To resolve the labelling problem that had persisted for centuries, Davies partnered with Professor Hassan Ugail, head of the Centre for Visual Computing at the University of Bradford. Ugail had already demonstrated the potential of artificial intelligence in art historical scholarship, having previously used his AI model to attribute a painting to Raphael that had puzzled experts for decades. His expertise in visual analysis and machine learning offered a promising avenue to reassess the Holbein corpus with fresh, objective scrutiny that human scholars alone might have overlooked.

The AI analysis demonstrated remarkably effective at recognising patterns beyond the reach of traditional connoisseurship. By analysing the entire corpus of Holbein sketches and assessing stylistic, compositional, and physiognomic features within the dataset, the researchers constructed a comprehensive system of visual relationships. This computational approach moved beyond subjective interpretation, instead exposing clusters of works that exhibited measurable similarities. The results challenged traditional assumptions and suggested that centuries of misattribution could finally be corrected through technological innovation.

The Framework Strategy

Ugail detailed the technique behind their pioneering examination: the team carefully assessed each image against every other piece in the Holbein corpus, building an vast visual database. The AI algorithm then pinpointed which sketches demonstrated the strongest stylistic and visual connections, clustering similar works together. This algorithmic mapping revealed unforeseen groupings that contradicted established classification, indicating that the eighteenth-century inscriptions had substantially misorganised the collection.

The results were remarkable. The unidentified woman sketch aligned closely with other confirmed portraits of the Boleyn-Howard family, whilst the Windsor sketch seemed to display stronger visual similarity with portraits of Elizabeth Howard, Anne Boleyn’s parent. These results pointed to a intentional swap of identities, revealing that misattribution had occurred at a specific moment when the collection was reorganised and renamed long ago.

The Windsor Sketch and Its Complex Issues

The Windsor sketch, one of two works at the centre of this re-evaluation, has persistently posed art historians with a puzzle that defied conventional wisdom. For centuries, it was confidently identified as Anne Boleyn, the ill-fated second wife of Henry VIII, whose dramatic fall from grace and death in 1536 have fascinated scholars and the public alike. Yet the physical characteristics depicted in the sketch seemed at odds with contemporary descriptions of Boleyn herself. The sitter is rendered in profile, revealing notably pale complexion and vivid red locks—features that appeared inconsistent with historical accounts of Boleyn’s deeper skin tone and different colouring.

Independent scholar Karen Davies began questioning this attribution whilst studying the wider Holbein corpus, which comprises more than 80 works covering the Renaissance master’s productive career. Her doubts were bolstered by the finding that the collection suffered from extensive labelling inconsistencies. A especially striking example concerned Henry Howard, the Earl of Surrey, whose portrait had been confused with that of his own father. These errors indicated that the 18th-century inscriptions, which had guided art historical understanding for generations, might be deeply questionable and in need of systematic re-examination.

Physical Features and Historical Documentation

Contemporary descriptions of Anne Boleyn, based on courtly accounts and historical documents, regularly depicted her as possessing a darker complexion than the figure shown in the Windsor sketch. Court observers recorded her striking appearance and bearing, yet the flame-haired, light-skinned woman depicted by Holbein seemed to diverge from these documented accounts. This discrepancy concerned Davies and initiated deeper investigation into whether the sketch’s long-accepted attribution had been incorrect from the beginning or whether it had been mislabelled during the eighteenth-century reorganisation of the Royal Collection.

The broader context of Holbein’s artistic practice and his careful attention to portraiture lends additional support for the reassessment. The Renaissance master was celebrated for his exacting observation and skilled technique, depicting subjects with exceptional precision. If the Windsor sketch genuinely depicts Elizabeth Howard, Boleyn’s mother, rather than Anne herself, it would mark a major reassessment to long-standing scholarly consensus. Such a revelation would necessitate detailed scrutiny of other works in the corpus, conceivably opening up a series of formerly overlooked identities and relationships within Tudor court circles.

Holbein’s Chaotic Years in the Court of the Tudors

Hans Holbein the Younger arrived in England at a time of particularly volatile periods, when Henry VIII’s marital upheavals and religious upheaval were reshaping the nation’s political landscape. The artist from Germany became the king’s favoured portraitist, recording the Tudor dynasty with unparalleled access and intimacy. Yet his standing in the royal household was anything but stable; he worked amidst perpetual scheming, executing portraits of figures who might lose favour—or face the executioner’s block—within a matter of weeks. His sketches captured not merely likenesses but fleeting moments of authority and fragility during a reign defined by sudden reversals of fortune.

The sketches in question were made throughout a time when Holbein’s work was most sought after, as Henry VIII aimed to commemorate his consecutive spouses and courtly companions. Producing with impressive speed and exactness, the master generated a large body of initial drawings and finished portraits that served as both creative records and political statements. His precise attention to detail and his skill in capturing the inner character of his models made his work essential to the royal establishment. Yet the very abundance of his output, alongside the turbulent conditions of the period, gave rise to misidentification and confusion in subsequent centuries.

• Holbein recorded Henry VIII’s court throughout one of its most turbulent political eras
• His access to royal subjects was unprecedented for a foreign-born artist
• Numerous individuals he portrayed were executed or exiled in the period following their depiction
• He created more than eighty paintings in his extensive Tudor court corpus
• His sketches served as both visual documentation and instruments of royal propaganda

Paving the Way for Future Research

The ramifications of this computational examination extend far beyond the two sketches in question. By establishing that algorithmic approaches can successfully challenge long-standing attributions and misidentifications, the research unlocks considerable potential for reassessing the entire Holbein corpus and similar collections located within major institutions. Karen Davies has been careful to frame her findings not as definitive conclusions but as invitations to scholarly debate, highlighting that the analysis generates inquiries rather than determining them conclusively. This judicious methodology embodies best practice in art historical study, where evidence needs to be carefully scrutinised and alternative interpretations reviewed before consensus can be reached.

The Royal Collection Trust has accepted the enduring doubt regarding the unnamed sitter’s origins and welcomed the innovative approach that AI technology contributes to the problem. As museums across the world contend with extensive collections of insufficiently catalogued period pieces, this undertaking shows how computational methods can act as a effective means for recovery and scholarly revision. Coming generations of experts may now employ similar AI methodologies to further contested pieces, possibly uncovering connections and identities that have been concealed for generations. The collaboration of traditional experts and digital specialists suggests a exciting new direction in Renaissance scholarship.