Three things to watch amid Anthropic’s latest feud with the government
Anthropic, an AI company based in the United States, developed an advanced AI model called Mythos, which demonstrated exceptional capabilities in handling code. Concerned about potential cybersecurity risks, Anthropic shared limited access with cybersecurity experts before releasing a modified version called Fable to the public. However, the U.S. government quickly intervened, citing national security concerns and imposing export restrictions on Fable. Anthropic subsequently revoked access to both models. This incident has sparked discussions about the broader implications of AI regulation, with some arguing that the government's response was overly reactive rather than well-planned. Additionally, the event has raised questions about the reliability of American AI companies, prompting European leaders to consider accelerating their own AI development efforts. Meanwhile, open-source AI models from China have gained attention due to their affordability and accessibility, potentially influencing global AI strategies.
The Trump administration has taken a significant step toward easing restrictions on Anthropic's powerful AI models following a two-week ban that disrupted their availability. On June 26, 2026, the administration announced that Anthropic could resume deploying its Mythos 5 model to over 100 U.S. government agencies and private sector companies. This decision marks a shift from earlier measures that had prohibited non-Americans, including Anthropic's own international staff, from accessing the models. The move comes after the administration reportedly found that adequate safeguards were in place to allow trusted partners to utilize the technology responsibly.
According to reports from Semafor and Reuters, the decision was made after the administration reviewed the security protocols surrounding the models. These protocols were initially deemed insufficient, leading to the temporary withdrawal of both Mythos 5 and Fable 5 from public access. However, the administration concluded that the necessary precautions were now in place, enabling the controlled distribution of Mythos 5 to select entities. The release of Fable 5 remains pending, as the administration has yet to determine whether it will follow a similar path.
The affected organizations include those responsible for safeguarding critical infrastructure, such as energy grids, transportation networks, and communication systems. Anthropic, which developed the models, expressed optimism about the outcome, stating that they were working closely with the U.S. government to restore full access to the models. In a statement posted on X, the company noted that the restoration process was progressing swiftly and that efforts would continue to expand access to Mythos 5 and eventually reintroduce Fable 5 for broader use.
The decision reflects a growing recognition of the importance of advanced AI tools in national security and cyber defense. Despite the administration's initial skepticism, the need for robust AI capabilities has become increasingly apparent, particularly in light of rising cyber threats. The National Security Agency (NSA) was among the institutions impacted by the ban, losing access to the models during the dispute. This situation highlighted the delicate balance between ensuring national security and fostering innovation in the AI sector.
The involvement of multiple stakeholders underscores the complexity of the issue. The Trump administration, represented by Commerce Secretary Howard Lutnick, played a central role in negotiating the terms of access. Anthropic, under the leadership of Chief Compute Officer Tom Brown, worked diligently to meet the regulatory requirements. Additionally, the U.S. government agencies and private sector firms that stand to benefit from the renewed access represent a diverse range of interests and operational needs.
Reactions from the involved parties have been largely positive, though some concerns remain regarding the long-term implications of the policy change. While the immediate focus is on restoring access to the models, there is ongoing discussion about how to ensure continued compliance with security standards. The administration has emphasized transparency, but questions persist about the extent of oversight and the potential risks associated with wider deployment.
Looking ahead, the administration is expected to continue monitoring the performance of the models and assessing the effectiveness of the safeguards in place. There is also anticipation that the release of Fable 5 may follow a similar trajectory, albeit with additional scrutiny due to its broader accessibility. As the landscape of AI continues to evolve, the interplay between regulation and technological advancement will remain a focal point for policymakers and industry leaders alike.
How each side covered it
The same event, grouped by the political lean of the outlets covering it.
progressive
center
conservative
★
How each side covered it
Support independent, bias-aware news and unlock the social pulse, community voting, and your personalized For You feed.
The United States has released a powerful AI model called Mythos developed by Anthropic, but access to this model is restricted to certain U.S. companies. This selective release suggests that the U.S. government is carefully managing the distribution of advanced artificial intelligence capabilities, potentially for national security or strategic advantage reasons. The move highlights growing global competition in AI development and underscores concerns over the potential misuse of such technology. By limiting access to specific firms, the U.S. aims to control the pace and direction of AI innovation within its borders while possibly restricting foreign entities from gaining similar advantages.
Bias read (Center): The article presents a factual report on the release of an AI model without overtly favoring any political perspective. It does not include biased language, one-sided sourcing, or omissions that would indicate a clear ideological leaning. The focus is on the technical and strategic aspects of the AI
Why these scores (Factual 100 · Objective 100): The article matches the primary source document exactly, stating that the US government has released the powerful Anthropic model Mythos to some US companies. It is factual and neutral in tone.
AxiosIndependentCenterFactual 100Objective 10013 days ago
The article discusses growing concerns over China's advancements in artificial intelligence (AI), particularly its development of open-source models such as GLM-5.2, which demonstrate capabilities comparable to leading U.S. models like Anthropic's Opus 4.8. Experts warn that these developments could accelerate cyber threats and challenge U.S. dominance in AI. However, there is debate over how quickly China is closing the gap, with some arguing that while Chinese models have improved significantly in quality, they may still lack the necessary computational resources and infrastructure to fully compete at the frontier level. Meanwhile, the U.S. faces internal disagreements over how to responsibly deploy its own advanced AI models, such as Anthropic's Fable 5 and Mythos 5, due to safety and national security concerns. Some experts caution against underestimating China's potential, while others emphasize the technical hurdles that remain.
Bias read (Center): The article presents multiple perspectives on China's AI advancements and their implications for U.S. security and technological leadership. It includes quotes from various experts, both highlighting concerns about China's progress and acknowledging the technical challenges that remain. There is no顯
Why these scores (Factual 100 · Objective 100): The article accurately reflects the primary source document, reporting that the US government has allowed the release of Anthropic's Mythos 5 model to select US institutions. It is concise and maintains a neutral tone.
SemaforIndependentCenterFactual 100Objective 10019 days ago
At the G7 summit, artificial intelligence (AI) industry leaders have proposed the creation of a global forum aimed at establishing standardized guidelines for advanced AI models. The initiative seeks to address concerns related to the ethical deployment, safety, and international cooperation regarding AI technologies.
Bias read (Center): The article presents a neutral summary of an event where AI CEOs propose a global forum to G7 leaders without taking a stance on the proposal itself. It does not include any biased language, one-sided sourcing, or editorializing that would indicate a particular ideological leaning.
Why these scores (Factual 100 · Objective 100): This article accurately reports the event without adding unverified details, remains neutral in tone, and focuses on the factual aspects of the event.
TechCrunchIndependentConservativeFactual 95Objective 909 days ago
The Trump administration has allowed Anthropic to provide its advanced cybersecurity-focused AI model, Mythos 5, to over 100 U.S. government agencies and companies, including permitting non-American employees within those organizations to access it. This decision comes two weeks after a ban that restricted access to Mythos 5 and Fable 5 due to concerns about potential security risks. The administration reportedly deemed sufficient safeguards in place for 'trusted partners' to use Mythos 5, though Fable 5 remains unavailable for broader deployment. Anthropic confirmed the progress in a statement on X, noting ongoing efforts to expand access to Mythos 5 and reintroduce Fable 5 for general use.
Bias read (Conservative): The article frames the administration's actions as a reversal of a prior restrictive policy, emphasizing the approval of access to a powerful AI model under conditions deemed secure by the government. The focus on 'trusted partners' and the selective nature of the access suggest a pro-government, or
Why these scores (Factual 95 · Objective 90): The article provides accurate information about the lifting of the block on Mythos 5 and mentions the exclusion of Fable 5. However, it slightly leans towards emphasizing the conflict with the Trump administration.
MIT Technology ReviewIndependentCenterFactual 85Objective 8014 days ago
Anthropic, an AI company based in the United States, developed an advanced AI model called Mythos, which demonstrated exceptional capabilities in handling code. Concerned about potential cybersecurity risks, Anthropic shared limited access with cybersecurity experts before releasing a modified version called Fable to the public. However, the U.S. government quickly intervened, citing national security concerns and imposing export restrictions on Fable. Anthropic subsequently revoked access to both models. This incident has sparked discussions about the broader implications of AI regulation, with some arguing that the government's response was overly reactive rather than well-planned. Additionally, the event has raised questions about the reliability of American AI companies, prompting European leaders to consider accelerating their own AI development efforts. Meanwhile, open-source AI models from China have gained attention due to their affordability and accessibility, potentially influencing global AI strategies.
Bias read (Center): The article presents a balanced view of the situation, discussing both the government's actions and the responses from Anthropic and international stakeholders. It does not exhibit strong bias toward either side, providing context and multiple perspectives without overtly favoring one over the other
Why these scores (Factual 85 · Objective 80): The article provides detailed background on the dispute between Anthropic and the government, including the initial release of Mythos and the subsequent ban. It offers a critical perspective on the government's response but remains mostly factual and balanced.
TechCrunchIndependentCenterFactual 85Objective 8017 days ago
The U.S. government recently imposed export restrictions on Anthropic's advanced AI models, Fable and Mythos, citing national security concerns. This move followed reports that a South Korean telecommunications firm linked to China gained access to Mythos, and claims that Amazon researchers bypassed Fable 5's security measures. Anthropic quickly restricted access to these models, which remained unavailable for a week. This incident highlights the challenges of using export controls to regulate emerging technologies like AI, similar to past attempts with encryption tools such as PGP in the 1990s. While the U.S. sought to limit PGP's spread due to fears it would hinder surveillance, the technology ultimately became widely adopted despite the restrictions.
Bias read (Center): The article presents the situation objectively, discussing both the government's actions and Anthropic's response without overtly favoring either side. It provides historical context regarding previous export control failures and includes perspectives from multiple stakeholders, including Anthropic,
Why these scores (Factual 85 · Objective 80): The article provides a comprehensive overview of the export control incident involving Anthropic's models. It includes relevant context and quotes from various sources while maintaining a relatively neutral stance.
AxiosIndependentCenterFactual 85Objective 8017 days ago
In an exclusive interview with 'The Axios Show,' former President Donald Trump discussed his previous concerns about Anthropic, an artificial intelligence company, being a national security threat. He indicated that while he once viewed the company as a potential risk, relations have since improved. The Trump administration imposed restrictions on access to Anthropic's advanced AI models, citing concerns raised by Amazon about vulnerabilities in the technology. Trump mentioned that Anthropic responded responsibly to these concerns, avoiding further escalation. While he did not rule out using emergency powers under the Defense Production Act, he emphasized that the U.S. is leading in AI development and wants to ensure it maintains an advantage over China.
Bias read (Center): The article presents Trump's statements neutrally, quoting him directly without overtly favoring one side. It provides context about the administration's actions and Anthropic's response, offering a balanced perspective on the situation without apparent ideological bias.
Why these scores (Factual 85 · Objective 80): The article offers insight into the internal workings of the Trump administration regarding AI policy. It remains largely objective while providing relevant contextual information.
Bloomberg NewsIndependent🔒CenterFactual 85Objective 8018 days ago
Anthropic PBC co-founder Jack Clark stated the company is engaging in daily discussions with the Trump administration to address security concerns related to its top AI models. He mentioned these efforts aim to align with the administration's goals of addressing national security risks posed by advancements in AI technology.
Bias read (Center): The article presents factual information without overtly favoring any political side. It reports on Anthropic's engagement with the Trump administration regarding AI model security without using biased language or emphasizing one perspective over another.
Why these scores (Factual 85 · Objective 80): The article accurately describes Anthropic's efforts to ensure AI safety and their ongoing discussions with the government. It maintains a relatively neutral tone.
AxiosIndependentCenterFactual 85Objective 7018 days ago
The article discusses the evolving leadership within the Trump administration's approach to artificial intelligence (AI) policy. Key figures such as David Sacks and Siriam Krishnan have stepped back or are leaving their roles, leading to a shift in influence toward a broader group of officials. Commerce Secretary Howard Lutnick has taken a prominent role, including imposing export controls on Anthropic and participating in international discussions on AI standards. Meanwhile, Chris Fall is handling technical aspects of AI policy domestically.
Bias read (Center): The article provides a factual overview of personnel changes and policy developments within the Trump administration's AI strategy without overtly favoring any political side. It reports on actions taken by officials like Howard Lutnick and mentions their roles and decisions without evaluative or sl
Why these scores (Factual 85 · Objective 70): Factual accuracy is strong regarding the Trump administration's AI policies and the roles of OpenAI and Anthropic. However, the article leans toward reporting on the White House's internal dynamics and shifts in leadership, which introduces some subjective framing. Objectivity is compromised by focu
AxiosIndependentCenterFactual 85Objective 6519 days ago
The White House has imposed export controls on Anthropic's latest AI model, raising concerns about the impact on the U.S. AI industry. Industry experts warn that such restrictions could hinder the financial growth of major AI companies like Anthropic and OpenAI, which rely on global adoption of their technologies. Analysts suggest that ongoing government interventions could disrupt business strategies and investment plans in the AI sector.
Bias read (Center): The article presents expert opinions and economic implications without overtly favoring any political side. It discusses concerns raised by industry analysts and researchers but does not take a stance on the policy itself or its justification.
Why these scores (Factual 85 · Objective 65): The Axios article accurately reports the export control issue but focuses on the broader implications for the AI industry rather than the specific details of the unblocking. It includes quotes from experts but leans towards expressing concern over the impact on U.S. tech firms, which slightly reduce
PolitiFactIndependentCenterFactual 80Objective 7512 days ago
At the GlobalFact 2026 conference in Vilnius, Lithuania, fact-checkers discussed the growing challenges they face, including political attacks, funding cuts, and the rise of AI-generated content undermining traditional fact-checking methods. Speakers like Angie Drobnic Holan and Neil Brown emphasized the importance of fact-checking as a defense against misinformation and repression. Nina Jankowicz, former member of the U.S. Disinformation Governance Board, criticized efforts to weaken anti-misinformation initiatives globally, citing actions by the Trump administration, Russian, and Chinese actors. She called for greater courage and emotional engagement in countering disinformation while maintaining ethical standards.
Bias read (Center): The article presents a balanced view of the challenges faced by fact-checkers without overtly favoring any political ideology. It highlights criticisms from multiple geopolitical actors (U.S., Russia, China) and includes perspectives from diverse stakeholders (international organizations, U.S. and欧盟
Why these scores (Factual 80 · Objective 75): The article provides accurate information about Trump's views on Anthropic but frames the narrative around his statements rather than presenting a balanced account of the situation.
TechCrunchIndependentCenterFactual 80Objective 7515 days ago
The Trump administration ordered Anthropic to take its two newest AI models, Fable 5 and Mythos 5, offline citing 'national security concerns,' though specific reasons were not disclosed. The move sparked debate over AI regulation and digital sovereignty. Anthropic complied, removing the models from public access, citing difficulties in determining if users were foreign nationals. The decision reportedly followed concerns raised by Amazon CEO Andy Jassy about potential vulnerabilities in Fable 5's safety mechanisms. Experts criticized the order, arguing it undermines U.S. cybersecurity by limiting access to advanced tools. Some suggested the situation could boost Anthropic's public image, portraying it as a 'bad boy' resisting government pressure.
Bias read (Center): The article presents multiple perspectives, including criticism from cybersecurity experts, comments from industry insiders, and contextual information about the administration's actions. It does not exhibit overtly biased language or one-sided sourcing, maintaining a balanced tone throughout the报道.
Why these scores (Factual 80 · Objective 75): The article discusses the export control order from the Trump administration and its impact on Anthropic's models. While it provides useful context, it lacks specific details about the unblocking of Mythos 5 and has a slightly opinionated tone regarding the potential benefits for Anthropic.
Bloomberg NewsIndependent🔒CenterFactual 80Objective 7519 days ago
Anthropic PBC has shut down two of its top AI models following pressure from the Trump administration, which raised national security concerns. Despite multiple meetings between Anthropic and Trump officials, no resolution has been reached.
Bias read (Center): The article presents facts without overtly favoring either side. It reports on an ongoing situation involving a tech firm and government action but does not use loaded language or emphasize one perspective over another.
Why these scores (Factual 80 · Objective 75): The article provides relevant background on the dispute but doesn't offer detailed updates on the current status of the export restrictions. Its neutrality is somewhat compromised by the framing of the issue.
QuartzIndependentCenterFactual 80Objective 7519 days ago
Trump stated that discussions with Anthropic are proceeding smoothly, despite the company having taken down its powerful AI models, Fable 5 and Mythos 5, following an administration ban on foreign access.
Bias read (Center): The article presents factual information without overtly favoring any political perspective. It reports on Trump's comments and the actions taken by Anthropic without using biased language or selectively presenting information.
Why these scores (Factual 80 · Objective 75): The article accurately reports the ban on foreign access to Anthropic's models and their shutdown. However, it omits the recent unblocking of Mythos 5 mentioned in the primary source. The tone leans slightly towards emphasizing the negative impact of the ban.
Bloomberg NewsIndependent🔒CenterFactual 75Objective 8519 days ago
Tech leaders, including CEOs from OpenAI and Anthropic, participated in discussions with G7 leaders in France regarding artificial intelligence. This follows concerns among US allies over the Trump administration's restrictions on access to Anthropic's advanced AI models.
Bias read (Center): The article presents a factual summary of events without overtly favoring any political side. It mentions the involvement of tech leaders and G7 discussions but does not provide commentary or framing that suggests a particular ideological stance.
Why these scores (Factual 75 · Objective 85): The article mentions the G7 meeting and US curbs but lacks specific details about the Mythos 5 unblocking. It references the broader context but doesn't align closely with the primary source document, which focuses on the US lifting the block on Mythos 5. The article is generally neutral.
Christian Science MonitorParty-alignedCenterFactual 75Objective 7018 days ago
Anthropic, a major artificial intelligence company, is engaged in a dispute with the U.S. government over export controls on its latest AI models. The conflict arises from concerns that these models might be used by foreign entities to launch cyberattacks against the United States. While the government seeks to impose restrictions for national security reasons, Anthropic argues that such measures hinder innovation. The situation underscores the complex relationship between AI developers and regulators, as both parties rely on each other despite their disagreements.
Bias read (Center): The article presents the dispute between Anthropic and the U.S. government in a balanced manner, without overtly favoring either side. It outlines the perspectives of both parties—Anthropic's concern about regulatory burdens and the government's emphasis on national security—without editorializing.
Why these scores (Factual 75 · Objective 70): The article gives a general overview of the situation but lacks specific details about the current status of Mythos 5 and Fable 5. It presents a balanced view but could provide more factual depth.
TechCrunchIndependentCenterFactual 70Objective 6017 days ago
The U.S. government recently compelled Anthropic to remove its two most recent AI models, Fable 5 and Mythos 5, due to national security concerns. This decision followed reports that Amazon researchers discovered a method to circumvent Fable 5's safety measures. Cybersecurity experts have raised alarms about the potential risks of such a ban, while Anthropic has pointed out similar vulnerabilities in other models. The situation has sparked debate over whether this action stems from legitimate security threats or reflects ongoing tensions with the Trump administration. The discussion also explores how this development could inadvertently benefit Anthropic, particularly regarding its upcoming initial public offering.
Bias read (Center): The article presents both perspectives—government actions based on national security concerns and criticisms from cybersecurity experts—without overtly favoring either side. It frames the issue as a debate rather than taking a clear stance.
Why these scores (Factual 70 · Objective 60): The article discusses the ban and potential implications but lacks specific details about the current status of Mythos 5 and Fable 5. It presents a somewhat biased perspective on the situation.
The New York Times (World)Independent🔒CenterFactual 65Objective 8019 days ago
Tech executives from Anthropic, OpenAI, and Mistral are set to attend a lunch meeting with G7 leaders.
Bias read (Center): The article reports on an event involving technology executives and world leaders without taking a stance or using biased language. It simply states the fact that these individuals will attend a meeting, without emphasizing any particular viewpoint or outcome.
Why these scores (Factual 65 · Objective 80): The article discusses tech CEOs meeting with G7 leaders but doesn't mention the unblocking of Mythos 5. It is neutral in tone but lacks the specific information presented in the primary source document.
The New York Times (US)Independent🔒CenterFactual 60Objective 7013 days ago
The article discusses an incident where the National Security Agency (N.S.A.) lost access to a powerful artificial intelligence model during a dispute with Anthropic, a major U.S. AI developer. This occurred amid the Trump administration's growing dependence on advanced AI systems for cybersecurity purposes. The situation highlights potential tensions between government agencies and private technology firms regarding access to critical AI tools. The loss of access could impact the N.S.A.'s ability to leverage cutting-edge AI for national security tasks.
Bias read (Center): The article presents a factual account of the N.S.A.'s loss of access to an AI model during a dispute with Anthropic, without overtly favoring either side. It mentions the Trump administration's reliance on AI for cybersecurity but does not take a stance on the dispute itself or imply blame. The phr
Why these scores (Factual 60 · Objective 70): The article mentions the NSA losing access to Anthropic's AI during the dispute but lacks specific details about the unblocking of Mythos 5. It is somewhat factual but doesn't fully align with the primary source document and presents a partial view of the situation.
TechCrunchIndependentCenterFactual 60Objective 559 days ago
Asian AI companies are releasing new large-scale AI models amid ongoing U.S. export restrictions on Anthropic's Mythos and Fable series. Chinese cybersecurity firm 360 announced Tulongfeng, claiming it can compete with Mythos, while Japanese startup Sakana AI launched Fugu, positioning it as comparable to Fable 5 and Mythos Preview. Both companies highlight the impact of the U.S. export ban, which limits access to these models for non-Americans. Sakana AI emphasized that its development of Fugu was intentional but timed to coincide with the export restrictions, noting that U.S. models still hold importance in Asia. Co-founder Ren Ito expressed support for maintaining open access to AI technologies and called for collaborative development.
Bias read (Center): The article presents a balanced perspective on the geopolitical implications of U.S. export bans on AI models, highlighting both the strategic moves by Asian firms and the continued relevance of U.S. AI in the region. While it discusses the broader implications of export control policies, it does so
Why these scores (Factual 60 · Objective 55): The article discusses unrelated topics and does not focus on the event in question. It lacks relevance and factual support related to the main event.
★
Keep the news honest.
ObjectiveNews is reader-funded and ad-free — we show you the bias instead of hiding it. Support independent journalism for €5/month.
United States
