New York Times chief slams AI companies
for 'theft' of intellectual property
The head of The New York Times newspaper launched a scathing attack on artificial intelligence companies, accusing them of "shameless theft of intellectual property" and of threatening journalism. He was speaking at a global media congress on Monday in the southern French city of Marseille.
Issued on: 02/06/2026 - RFI
According to Arthur Gregg Sulzberger, president and publisher of the prestigious The New York Times newspaper, "tech giants strip-mine news websites without permission or compensation."
He was addressing newspaper and news site executives from around the world gathered on the first of a three-day World News Media Congress.
Sulzberger accused artificial intelligence companies' "hijacking of the public square is made possible by the original sin that animates their AI products – a brazen theft of intellectual property that has occurred at an unprecedented scale."
"They reclaim this stolen content as if they were its authors, thus diverting the audience and revenue" of news sites," he declared in a speech that was met with enthusiastic applause.
He added that the news sector "has been too quiet, too passive and too fragmented in the face of abuses by the companies leading the AI revolution."
Fragile business model
While The New York Times a strong presence in the American press with more than 13 million subscribers to its print and digital products, thanks in particular to strong diversification (podcasts, cooking recipes), the United States has lost up to 3,000 newspapers in the last two decades, Sulzberger recalled.
"I fear we are careening toward a future with fewer and fewer journalists to do the expensive, difficult work of original reporting," he said.
Already facing competition from social media, traditional media outlets are seeing their fragile business model put under pressure by chatbots like ChatGPT (OpenAI).
French press take on digital databases to defend journalist copyright against AI
These chatbots respond directly to user requests, thanks to AI models that feed on the internet, particularly press content, consequently reducing traffic to news sites.
Sulzberger, whose company is suing OpenAI, the creator of ChatGPT and Microsoft over the use of copyrighted work, said AI companies are "consolidating their outsize control over our data and our attention" but are "failing to embrace a core responsibility that comes with this power – to ensure the public has access to trustworthy news and information."
The legal battle that has lasted two and a half years and has already cost the thriving American newspaper "more than $20 million" (€17 million) in legal fees.
"AI companies know this: most media outlets don't have the resources to go to court to enforce their rights," he added.
Crucial moment in media history
After the arrival of the internet and social media, "we are once again at a crucial moment in the history of media," Ladina Heimgartner, president of WAN-IFRA and CEO of the Swiss media group Ringier Media, told French news agency AFP.
"We are experiencing an unacceptable loss of value," but we can only "move forward by engaging in dialogue with tech companies," she added.
The tech giant Google, one of the first to reach agreements to pay news publishers for their content, is a partner of the conference, co-organised by CMA Media, the media arm of the French shipping company CMA-CGM.
AI assistants 'not reliable' when it comes to news, major European study finds
Representatives from OpenAI have also been invited. For Jean-Christophe Tortora, CEO of CMA Media, a "new deal between publishers, tech platforms, and the government" is needed to address the threat to the "sustainability" of the press.
"Those who think we can get away with it by each making deals on our own with one or two platforms, that's a short-term bet," he warned, while media outlets and AI players have already concluded compensation agreements.
In France, Apig, a collective body representing nearly 300 French daily newspapers, announced on Monday that it is suing the Californian company Brave, which operates an internet browser and search engine, accusing it of plundering its members' content using artificial intelligence.
(with AFP)
Microsoft launches its own AI models to take on OpenAI and Anthropic
Seven in-house models unveiled at Build 2026 signal Microsoft's push to cut costs and compete directly on the AI frontier as its biggest investees plan to launch record-breaking IPOs.
Microsoft has unveiled a family of seven in-house AI models at its annual Build developer conference in San Francisco, in the clearest sign yet that the tech giant is moving to reduce its dependence on the AI companies it has poured billions into.
Mustafa Suleyman, Microsoft AI chief executive, said that after tuning its models for consulting firm McKinsey, the company outperformed OpenAI's GPT-5.5 on quality — with what it projects as ten times better cost efficiency, based on public pricing data scaled across model sizes.
"We believe the time has come for every company to move from consuming a frontier model to fully participating at the frontier," said chief executive Satya Nadella at the conference.
The headline release is MAI-Thinking-1, Microsoft's first reasoning model, trained from scratch on clean, commercially licensed data without distillation from third-party systems.
A mid-sized model with 35 billion active parameters and a 256,000-token context window, it is designed for complex multi-step instructions, long-context reasoning and code generation.
Alongside it, the company launched MAI-Code-1-Flash, a coding model that converts text descriptions into source code for applications and websites, now rolling out across GitHub Copilot and Visual Studio Code.
By running its own models on Azure infrastructure, Microsoft can sidestep the fees it currently pays to third-party providers — and pass the savings to developers.
According to Microsoft, in blind evaluations run by Surge, its independent human rating partner, MAI-Thinking-1 was preferred over Anthropic's Claude Sonnet 4.6, and the company says it matches Claude Opus 4.6 on coding benchmarks.
Quantum leap
Also at Build, Microsoft announced that its Majorana 2 quantum chip is 1,000 times more reliable than its predecessor — a milestone the company said brings it within striking distance of a commercially useful quantum computer.
Qubits, the fundamental unit of quantum computing, are notoriously fragile: even minor temperature shifts or vibrations can knock them off course. The Majorana 2 chip addresses this directly.
Qubits on the new chip survive for an average of 20 seconds, compared to milliseconds on the original, an improvement the company likened to upgrading from a phone that needs daily charging to one that lasts several years.
"We will have a quantum machine in 2029 that can solve commercially viable, reasonable problems," said Zulfi Alam, corporate vice president of Microsoft Quantum. The chip currently has just 12 qubits, while a useful machine would require millions.
Microsoft's approach centres on so-called topological qubits, based on the properties of a quasi-particle first theorised in the 1930s by Italian physicist Ettore Majorana.
The path has not been smooth — the company was forced to retract a 2018 paper in the journal Nature claiming evidence for the particle — but it has pressed on, and the second-generation chip improves on the first partly by replacing aluminium with lead as a superconductor.
The chip and its supporting research have not yet been peer reviewed, and some physicists have called for more information.
IPO race heats up
Microsoft's push for model independence comes as the companies it has committed billions to prepare for blockbuster stock market debuts.
Anthropic, the AI lab behind Claude, filed confidentially for an initial public offering on 1 June, just days after raising $65 billion (€59bn) in a Series H funding round that pushed its valuation to $965 billion (€877bn). OpenAI is also readying its own confidential filing.
Microsoft has committed $13 billion (€11.8bn) to OpenAI and invested up to $5 billion (€4.5bn) in Anthropic, while making both companies' models available through Azure.
June 3, 2026
By Deborah Barsky
Each leap in human communication—from vocal anatomy to writing to digital networks—followed the same pattern: faster, more complex, less individual.
Many people are overwhelmed by the fast-paced evolution of mass communication in a world increasingly shaped by the internet and artificial intelligence (AI). Yet ideas have not always circulated across the globe at lightning speed.
Looking into deep time allows us to view our current mode of existence from a broader perspective and to discern patterns of change that may offer insights into our species’ possible trajectories in the future. We can begin by trying to identify the conditions under which humans developed complex symbolic communication, which is considered unique in the animal kingdom. Today, we have succeeded in transposing incredibly complex digital languages, beyond the grasp of the human mind, into computerized systems capable of processing, storing, and sharing all kinds of information with the push of a button or a click of a mouse.
Human language is at the root of this system. While hardly perceptible in the archeological record, hominin physiognomy assessed from the fossil record allows us to observe that the capacity to vocalize across a significant range dates back to a very early time period. Physical features linked to resonance complexity and articular precision, like the low positioning of the larynx, an enlarged pharyngeal cavity, and a flexible tongue housed in an ample palate, were already part of the human anatomy by the time Homo erectus emerged in Africa nearly 2 million years ago.
Somatic harmonization, necessary for speech production, is ensured by the tiny and crucial hyoid bone, a floating anchor for the muscles controlling the tongue, throat, and voice box as they work together to regulate sounds emitted during speech. Paleoanthropologists studying the position, geometry, and internal structure of Neanderthal hyoids discovered in archeological sites found them to be nearly identical to those of anatomically modern humans, suggesting that they were already capable of emitting a similarly sophisticated range of sounds. Evidence from Spain also reveals that modern hyoid morphology was present by at least 530,000 years ago, indicating that it could be a derived characteristic shared by modern humans and pre-Neandertals, which they inherited from their last common ancestor.
The ability to emit a wide range of vocalizations characteristic of modern human language did not appear abruptly, but rather, was favored over millennia by interrelated adaptive elections taking place under the evolutionary orchestration of both natural selection and techno selection. As early as around 7 million years ago, some early primates displayed several craniofacial modifications believed to have evolved, at least in part, to accommodate bipedal locomotion. These included features such as reduced prognathism, more gracile jawbones with smaller teeth, and a restructured skull base, all of which subsequently facilitated the anatomical tweaks necessary to increase the human capacity to pronounce a wide range of sounds.
By the time hominins began systematically making stone tools some 2.6 million years ago, a series of changes in the archeological record reflected the significance of this adaptive strategy. Around 1.75 million years ago in Africa, these (Oldowan) assemblages were progressively replaced by Acheulian industries comprising tools with wider technological and morphological variability, including skillfully manufactured and (esthetically pleasing) premeditated forms, like spheres and symmetrical pointed tools called handaxes, which required high degrees of dexterity to make.
Observing synapses activated during experimental tool replication has shown that the cognitive processes involved in passing on the skills necessary to make such tools are linked to those activated by language. Furthermore, evidence from ancient DNA analysis suggests that regulatory genetic variants favoring language-producing abilities underwent long-term evolutionary selection in hominin populations, with some of these features predating the divergence of modern humans and Neanderthals.
Fully embracing the advantages offered by their newfound technological proficiency, Acheulian hominins benefited from access to the choicest morsels of meat and viscera to feed their developing brains. They experienced unprecedented demographic expansion and assembled into more complex group settings sheltering numerous individuals. Eventually wielding fire, Acheulian populations spread into new lands within and beyond Africa, utilizing their rapidly developing technologies to confront diverse environmental and social challenges.
Some archeologists have described this process as a biocultural feedback loop, in which unprecedented cerebral growth (supported by higher protein intake) was linked to toolmaking. This process was catalyzed by increasing social complexity, reinforced by the development of culture, and favored the selection of symbolic communication. Language—the preponderant expression of symbolic transmission—initially developed as a survival-related response to effectively transmit technological competence, and ultimately became a means to express multifaceted cultural norms developed to promote social proficiency.
There is no doubt that culture evolves; that it was/is transmitted socially through learning processes that have been intricately developed over time and now extend well past adolescence and, in some cases, beyond the optimal reproductive age for women. Culture is inherently cumulative: it is an aggregate of past human experiences, errors, discoveries, adaptations, and rectifications to deduce scientific and philosophical insights about the universe—and our place in it. Symbolic communication, expressed as language and writing, and later also as art and music, reinforces cumulative culture by allowing us to express, preserve, and share knowledge beyond time constraints.
The story of technological evolution following the emergence of the Acheulian reveals two major forces that still operate today: acceleration and complexification. By the end of the Acheulian and into the Middle Paleolithic period and beyond, technological and (linked) social achievements accumulated exponentially, widening the baseline upon which hugely transformative accomplishments emerged. At the same time, populations multiplied, creating interrelational networks that favored reproductive, material, and cultural exchanges.
I have proposed defining this phenomenon as a fractalization process, insofar as each branch replicates the underlying evolutionary patterns, overcoming the constraints of its ancestral source while retaining the full spectrum of human cognitive expression. In this model (that is comparable to biological natural selection), small variations or adjustments can lead to exponentially magnified outcomes through recursive and effectively infinite cycles of reproduction.
By the time our own species came on the scene more than 200,000 years ago, culture had flourished into symbolic material expressions, inexorably linked to the technological and social realities of past human lives. Fire, for example, provided ample fodder for experimenting with the power of transforming existing materials, enabling new capacities during the Neolithic and Metal Ages, with successive discoveries, such as pottery and, eventually, glass and metallurgy, transmuting and enabling new societal paradigms.
Following the shift from hunter-gatherer to production-based lifeways, technological complexity split individual role-playing into distinct occupations, sharpening perceptions about (imagined) interpersonal “value” within social settings. As populations swelled and collective living arrangements expanded, these systems became structured around complex symbolic interactions, material cultures, territorial identities, and hierarchies constructed to sustain group cohesion. Importantly, social organization came to rely upon artificially constructed (symbolic) notions that led regionally assembled groups to contrast their ideas of communal belonging (identity) with those existing in neighboring territories. As competition for land and resources increased, these contrasting symbolic realities became a justification for (sometimes violent) interpopulation conflicts.
Communication networks grew increasingly complex alongside expanding production and territorial markets, accelerating the spread of cultural and material exchanges. Protohistoric societies developed early writing systems based on symbolic signs used as mnemonic strategies to represent ideas and objects. Emerging independently in regions such as Mesopotamia, Egypt, Central America, and China, these systems gradually evolved from pictorial symbols into stylized phonetic and logographic writing largely detached from their original visual meanings.
Throughout human experience, learning has been central to survival, with scientific and behavioral complexification acting as a driving force to develop sophisticated new ways to educate, demonstrate, and train each successive generation to become effective agents of our seemingly unstoppable technological proficiency. Presently, ideas move instantaneously across space and time through global communication networks. Digital platforms make information accessible worldwide, shrinking distances and connecting populations, while promoting mass media and consumer culture. Often shaped by corporate interests that influence production, consumption, and resource distribution in a world driven by digital images and symbolic messaging, global connectivity has become the force shaping human thought and social organization.
Uniformity is increasingly becoming a defining characteristic of human life shaped by digital communication. Despite the technological sophistication of the systems that mediate and transmit human thought, we are adopting simpler forms of communication, homogenizing language, and gradually adapting our patterns of thinking to fit the digital platforms used to express them. The next stage of human evolution involving the merging of human intelligence with AI may already be here. As we become increasingly dependent on computerized tools to function within contemporary social environments, the ability to communicate effectively through digital means is emerging as a fundamental requirement.
At the end of the 1980s, screenwriter Maurice Hurley and the creators of Star Trek showed remarkable foresight in creating the Borg: fictional cybernetic organisms linked through a single collective consciousness in which thoughts and actions are shared and controlled by a powerful centralized computer system. In this dystopian vision of the future, the Borg’s main purpose is to assimilate other species in pursuit of greater knowledge, efficiency, and power. Once assimilated, subjects lose their individuality and become linked to the Collective (or hive mind).
As humanity moves toward a predicted state of post-humanism (human-machine hybridization) and the boundaries between human cognition and machine intelligence become increasingly blurred, it is worth asking whether this vision already mirrors some aspects of our own interconnected and technology-mediated world.
Author Bio:
Credit Line: This article was produced by Human Bridges.
No comments:
Post a Comment