Home Blog Page 194

Cloudflare Outage: How a ‘latent bug’ triggered a global disruption of X, ChatGPT services | Explained

0

Several of the world’s largest online services, including X, ChatGPT, and numerous websites that depend on Cloudflare for security and traffic routing, were disrupted on November 18, 2025, when a significant outage rippled across the internet. Users reported slow loading, broken pages, and complete downtime for platforms that typically handle billions of daily requests. As confusion mounted, Cloudflare’s Chief Technology Officer, Dane Knecht, posted a detailed explanation on Twitter, outlining the internal failure that cascaded into a global disruption.

In his message, Mr. Knecht acknowledged that Cloudflare had “failed” its customers and the broader internet. He emphasised that organisations across the world rely on Cloudflare to keep their websites and applications accessible, and on this particular day, the company did not uphold that responsibility. What looked from the outside like a sudden, widespread network collapse stemmed from a highly technical but critical component inside Cloudflare’s infrastructure: its bot-mitigation system.

What is a bot-mitigation system?

To understand why the failure caused such extensive disruption, it’s essential to understand what a bot-mitigation system actually is. The modern internet is flooded with automated traffic. Not all bots behave maliciously. For instance, search engines, uptime monitors, and legitimate APIs rely on automated processes. But a significant portion of bots exist to cause harm or unfairly exploit online systems. These harmful bots attempt credential-stuffing attacks using leaked passwords, scrape websites to steal content or competitive information, test servers for potential security vulnerabilities, overwhelm sites with junk traffic, or otherwise distort normal usage.

Bot-mitigation systems exist to keep this type of abusive automated traffic away from websites and applications. Cloudflare’s system analyses vast amounts of web traffic in real time, using a combination of behavioural analysis, machine-learning models, network fingerprinting, challenge-response mechanisms, and IP-reputation tracking. It scrutinises how quickly a user, or a bot, moves between pages, whether headers match known browser patterns, whether traffic resembles human interactions, and how the request compares to global patterns across millions of clients. Many of these checks are invisible to normal users, but they play an essential role in preventing everything from data theft to full-blown outages caused by bot overload.

Does only Cloudflare use bot-mitigation systems?

Cloudflare is not unique in running such systems. Virtually every major infrastructure provider that handles web traffic at scale has its own bot-mitigation architecture. Amazon Web Services, Google Cloud, and other content-delivery providers maintain similar systems that separate harmful and legitimate traffic before it reaches the websites using their services. Without these layers of automated protection, the modern internet would be much more fragile, susceptible to constant low-grade attacks, and significantly slower for everyday users.

What is a latent bug?

What made the incident particularly notable was that the flaw was what Mr. Knecht described as a “latent bug.” A latent bug is an error that sits hidden in the system, often for months or years, without causing any visible issues. These are among the most difficult flaws to detect because they remain dormant under everyday conditions. They often require a rare or unusual combination of inputs or environmental conditions to activate. Only when that specific combination occurs does the underlying flaw suddenly emerge and cause unpredictable, sometimes severe, effects.

In this case, the latent bug existed inside a service responsible for supporting Cloudflare’s bot-mitigation capabilities, per what the CTO posted on X. Under normal operations, the bug probably didn’t interfere with the system’s functioning. It remained silent until a specific configuration update created exactly the sequence of events needed to trigger the crash.

Once the service started failing repeatedly, the problem cascaded to other interconnected systems, leading to a broad degradation across Cloudflare’s network. Although the issue originated in a subsystem dedicated to handling automated traffic, the ripple effect reached far beyond that, affecting practically every service that depends on Cloudflare’s infrastructure.

Mr. Knecht emphasised that the disruption was not the result of an external attack. Instead, it was an internal systems failure exacerbated by the scale and interdependence of Cloudflare’s services. Many modern internet outages have similar root causes: an unexpected failure born from the complexity of distributed systems rather than malicious activity. When companies operate thousands of servers across hundreds of regions and handle an enormous share of global traffic, even small internal faults can create disproportionately large external consequences.

What is a routine configuration change?

The incident stemmed from what the CTO described as a “routine configuration change,” which is another key concept in understanding why outages like this occur. Large internet infrastructure providers regularly make configuration updates to keep systems running smoothly. These updates are not the same as rewriting software or deploying new code. Instead, they involve adjusting the internal parameters that define system behaviour. A typical routine update could involve modifying traffic-routing rules, updating threat-detection models, adjusting timeout or capacity settings, switching on new features, or updating lists of known malicious IP ranges.

Such updates occur constantly. They are considered safe because they usually pass through extensive automated testing, and companies roll them out in stages to reduce the risk of widespread disruption. However, even with these safeguards, the sheer complexity of global infrastructure means that unexpected interactions sometimes slip through. When a latent bug meets an ordinary update, the result can be a cascading failure, exactly the situation Cloudflare found itself managing.

In his message, the CTO noted that Cloudflare had already fixed the issue and the company is now working on long-term fixes to prevent the same flaw from reemerging. He also noted that more detailed information about the cause of this issue will be shared by the company.

This outage, which follows less than a month after the AWS outage, serves as a reminder of how interconnected the internet is and how much of it passes through infrastructure providers. It also illustrates the fragile balance between complexity and reliability that underpins the online world. A single bug, dormant and undetected, combined with an ordinary configuration change, can ripple across continents and disrupt services used by hundreds of millions of people.

Published – November 18, 2025 11:09 pm IST



Source link

No firm is immune if AI bubble bursts: Google CEO

0

 Alphabet Chief Executive Sundar Pichai said no company would be unscathed if the artificial intelligence boom collapses, as soaring valuations and heavy investment in the sector fuel concerns of a bubble.

Mr. Pichai said in an interview with the BBC published on Tuesday (November 18, 2025) that the current wave of AI investment was an “extraordinary moment” but acknowledged “elements of irrationality” in the market, echoing warnings of “irrational exuberance” during the dotcom era.

There has also been much debate among analysts about whether AI valuations are sustainable.

Asked about how Google would cope with a potential bursting of a bubble, Mr. Pichai said he thought it could weather the storm but added: “I think no company is going to be immune, including us.”

Alphabet shares have surged about 46% this year, as investors bet on its ability to compete with ChatGPT-maker OpenAI.

In the United States, concerns about lofty AI valuations have begun to weigh on broader markets, while British policymakers have also flagged bubble risks.

In September, Alphabet pledged 5 billion pounds over two years for UK AI infrastructure and research, including a new data centre and investment in DeepMind, its London-based AI lab.

Mr. Pichai also told the BBC in the interview conducted at Google’s California headquarters that Google would begin training models in Britain, a move Prime Minister Keir Starmer hopes will bolster the country’s ambition to be the world’s third AI “superpower” after the United States and China.

Mr. Pichai also warned of the “immense” energy needs of AI and said Alphabet’s net-zero targets would be delayed as it scales up computing power.

Opinions split over AI bubble

Here is a list of industry executives, economists, investors and analysts’ takes on the topic:

Morten Wierod, CEO OF ABB: “I don’t think there is a bubble, but we do see some constraints in terms of construction capacity not keeping up with all the new investments,” Wierod told Reuters on October 16.

“We are talking about trillions in investment,” he said. “That will take a few years to implement because there is not enough people and resources to build all this.”

Jeff Bezos, FOUNDER AND EXECUTIVE CHAIRMAN OF AMAZON: “When people get very excited as they are today about artificial intelligence, for example, every experiment gets funded … And investors have a hard time in the middle of this excitement distinguishing between the good ideas and the bad ideas,” Bezos said during the Italian tech week on October 3.

“A bubble like a banking bubble, a crisis in the banking system, that’s just bad … The ones that are industrial are not nearly as bad, it could even be good because when the dust settles and you see who are the winners, society benefits from those inventions.”

Bank of England: Global markets could tumble if investors’ mood sours on the prospects for AI, the Bank of England said on October 8.

“The risk of a sharp market correction has increased,” the BoE’s Financial Policy Committee said in a quarterly update, in its sharpest warning to date of the dangers of an AI-triggered market slump, adding that the risk of spillovers to Britain’s financial system from such a shock was “material”.

Bryan Yeo, CHIEF INVESTMENT OFFICER AT GIC: “There’s a little bit of a hype bubble going on in the early-stage venture space,” said Singapore sovereign wealth fund’s Yeo during a panel discussion at the Milken Institute Asia Summit on October 3.

“Any company startup with an AI label will be valued right up there at huge multiples of whatever the small revenue (is) … That might be fair for some companies and probably not for others.”

Joseph Briggs, ECONOMIST AT GOLDMAN SACHS’ GLOBAL ECONOMICS RESEARCH: The flood of multibillion-dollar investments pouring into U.S. AI infrastructure is sustainable, pushing back on mounting concerns that the sector’s spending spree could be overheating, Briggs said in a note on October 16.

While the overall macroeconomic case for AI investment remains strong, he cautioned that “the ultimate AI winners remain less clear”, with fast technological change and low switching costs potentially limiting first-mover advantages.

Pierre-Olivier Gourinchas, CHIEF ECONOMIST AT IMF: The AI investment boom in the U.S. may be followed by a dotcom-style bust, but it is less likely to be a systemic event that would crater the U.S. or global economy, Gourinchas said on October 14.

“This is not financed by debt, and that means that if there is a market correction, some shareholders, some equity holders, may lose out.”

Sam Altman, CEO OF OPENAI:

“Are we in a phase where investors as a whole are overexcited about AI? My answer is yes,” Altman told tech media The Verge in August.

“Someone is going to lose a phenomenal amount of money. We don’t know who, and a lot of people are going to make a phenomenal amount of money.”

Michael Burry, INVESTOR AND FOUNDER OF SCION ASSET MANAGEMENT: The “Big Short” investor has placed bearish bets on Nvidia and Palantir.

Last month, in his first X post in more than two years, Burry warned of a bubble, fanning investor concerns over inflated spending in the AI and tech industry.

UBS: Almost as many investors who feel we are in an AI bubble are also hanging on to their investments in the sector, UBS equity strategists said on October 14.

“Most felt we were in an AI bubble, but that far from the apex of a bubble peak and thus around 90% of the people who said we were in a bubble said they were still invested in many of the AI-related areas.”

Published – November 18, 2025 07:23 pm IST

Source link

Paciente evita la enfermedad de Lyme, pero recibe una factura sorpresa

0

Un fin de semana de fines de abril, Leah Kovitch estaba arrancando plantas invasoras en un prado cerca de su casa cuando una garrapata se le prendió en la pierna.

No notó al diminuto insecto hasta el lunes, cuando empezó a sentir dolor en el músculo de la pantorrilla. Esa misma mañana, hizo una cita virtual con una doctora —recomendada por su plan de salud—, quien le recetó un tratamiento de 10 días con doxiciclina para prevenir la enfermedad de Lyme, y le insistió en que también fuera a una consulta en persona. Así que, más tarde ese mismo día, fue, sin cita previa, a una clínica cerca de su casa en Brunswick, Maine, donde la evaluaron y le recetaron una dosis más alta, única, del mismo medicamento.

Fue una buena decisión porque el personal de la clínica encontró otra garrapata en el cuerpo de Leah durante esa visita. Además, después de enviar uno de los insectos a un laboratorio para su análisis, el resultado de la prueba fue positivo para Lyme.

“Pude haberme enfermado seriamente”, dijo Kovitch.

Pero la aseguradora de Kovitch rechazó cubrir la visita a la clínica. ¿La razón? No había obtenido una derivación de su médico para ver un especialista ni autorización previa. “Su plan no cubre este tipo de atención, por lo tanto, rechazamos este cargo”, explicaba un documento.

Las aseguradoras de salud han argumentado durante años que la autorización previa ayuda a reducir el fraude, los gastos innecesarios y a proteger a los pacientes. Y si bien estos rechazos suelen asociarse con tratamientos costosos, como los del cáncer, la diminuta picadura de una garrapata que sufrió Kovitch muestra cómo las compañías también utilizan esta política para evitar pagar por servicios de todo tipo, incluso cuando son considerados económicos y médicamente necesarios.

Promesas de cambio

El gobierno del presidente Donald Trump anunció este verano que docenas de aseguradoras privadas de salud aceptaron realizar cambios significativos en el proceso de autorización previa.

La promesa incluye eliminar por completo el requisito de autorización para ciertos servicios médicos. También se acordó otorgar un período de gracia a pacientes que cambian de plan médico, para que no enfrenten nuevas reglas que interrumpan sus tratamientos en curso.

Mehmet Oz, administrador de los Centros de Servicios de Medicare y Medicaid (CMS), anunció en una conferencia de prensa en junio que algunos de los cambios entrarían en vigencia en enero.

Pero, hasta ahora, el gobierno federal ha ofrecido pocos detalles sobre cuáles de los códigos de diagnóstico —utilizados para fines de facturación médica— quedarán exentos de autorización previa, o cómo hará que las aseguradoras privadas cumplan las nuevas reglas. No está claro si casos como el de Kovitch, relacionados con la enfermedad de Lyme, estarían exentos.

Chris Bond, vocero de AHIP, el principal grupo comercial de la industria de seguros médicos, confirmó que las aseguradoras se comprometieron a implementar algunos de los cambios para el 1 de enero. Otros cambios tomarán más tiempo. Por ejemplo, las compañías acordaron responder al 80% de las solicitudes de autorización en “tiempo real”, pero eso no ocurrirá sino hasta 2027.

Andrew Nixon, vocero del Departamento de Salud y Servicios Humanos de Estados Unidos (HHS), explicó a KFF Health News que los cambios prometidos por las aseguradoras buscan “reducir la burocracia, acelerar las decisiones de atención médica y fomentar la transparencia”, aunque advirtió que requerirán tiempo para lograr un impacto completo.

Mientras tanto, algunos expertos en políticas de salud se muestran escépticos sobre si las compañías realmente cumplirán con lo prometido. No es la primera vez que las grandes aseguradoras anuncian una reforma del proceso de autorización previa.

Luego de ir a una clínica sin cita previa por una picadura de garrapata, Kovitch descubrió que su aseguradora no cubriría el costo de la consulta porque, según ella, no había obtenido una derivación ni una autorización previa. Intentó apelar la decisión, sin éxito, y finalmente tuvo que pagar $238 de su propio bolsillo por la atención recibida en la clínica.(Brianna Soukup for KFF Health News)

Bobby Mukkamala, presidente de la American Medical Association (AMA), escribió en julio que las promesas hechas por las aseguradoras en junio son “casi idénticas” a las que la industria del seguro hizo en 2018.

“Creo que esto es una estafa”, opinó Neal Shah, autor del libro Insured to Edith: How Health Insurance Screws Over Americans — And How We Take It Back (“Asegurados hasta la muerte: cómo el seguro de salud perjudica a los estadounidenses y cómo podemos recuperarlo”).

Según Shah, las aseguradoras firmaron el acuerdo impulsadas por la presión pública. La indignación colectiva contra las compañías aseguradoras aumentó tras la muerte del director ejecutivo de United Healthcare, Brian Thompson, en diciembre. Oz indicó que el compromiso de las aseguradoras fue una respuesta a la “violencia en las calles”.

“Cada vez rechazan más reclamos”, dijo Shah, que es uno de los fundadores de Counterforce Health, una compañía que usa inteligencia artificial para ayudar a los pacientes a apelar las negativas del seguro. “Nadie se hace responsable.”

Resolver el caso

La factura que Kovitch recibió por su cita en la clínica fue de $238 y tuvo que pagarla de su bolsillo luego de enterarse de que su aseguradora, Anthem, no cubriría ni un centavo. Primero intentó apelar la decisión. Incluso consiguió una remisión retroactiva de su doctora de atención primaria, que respaldaba la necesidad de la visita.

No funcionó. Anthem volvió a negar la cobertura. Kovitch dijo que cuando llamó para averiguar la razón, la representante con la que habló no supo explicarle.

“Era como si no lo entendieran”, explicó Kovitch. “Todo lo que repetían, una y otra vez, era que no tenía autorización previa”.

Después, Jim Turner, vocero de Anthem, atribuyó el rechazo de la aseguradora a un “error de facturación” cometido por Maine Health, el sistema de salud que opera la clínica donde Kovitch fue atendida. Según Turner, el error provocó que el reclamo se procesara como si fuera una visita a un especialista, en lugar de una visita de atención sin cita previa o de urgencia.

Turner no proporcionó documentación que mostrara cómo ocurrió el error. Los registros médicos que Kovitch entregó muestran que Maine Health codificó su visita como “mordida de garrapata en la parte inferior izquierda de la pierna, primer encuentro”, y no queda claro por qué Anthem la interpretó como una visita a un especialista.

Después de que KFF Health News contactara a Anthem para preguntar sobre la factura de Kovitch, Turner dijo que la compañía “debió haber identificado el error de facturación antes, en el proceso, y pedimos disculpas por los inconvenientes que esto le causó a la señora Kovitch”.

Caroline Cornish, vocera de Maine Health, dijo que no es la primera vez que Anthem niega cobertura a pacientes que llegan sin cita previa. Señaló que las reglas de procesamiento de Anthem a veces se aplican de forma incorrecta a este tipo de visitas, lo que lleva a “rechazos inapropiados”.

Afirmó que estas visitas no deberían requerir autorización previa y que el caso de Kovitch ilustra cómo las aseguradoras suelen utilizar los rechazos administrativos como respuesta inicial.

“Maine Health considera que las aseguradoras deberían enfocarse en pagar la atención que sus afiliados necesitan, en lugar de crear obstáculos que retrasan la cobertura y pueden desalentar a los pacientes a buscar atención”, dijo. “El sistema, con demasiada frecuencia, está en contra de las personas a las que se supone que debe servir”, agregó.

Finalmente, en octubre, Anthem le envió a Kovitch una actualización de su resumen de beneficios, en la que se indicaba que una combinación de pagos de la aseguradora y descuentos cubriría el costo total de la consulta. Kovitch contó que una representante de la empresa la llamó para disculparse. A principios de noviembre, recibió el reembolso de los $238.

Pero hace poco se enteró de que, según nuevas reglas establecidas por Anthem, su cita anual con el oftalmólogo ahora requiere una derivación de su doctora de atención primaria.

“Esto sigue igual”, dijo. “Pero ahora conozco mejor cómo actúan”.



Source link

Peter Thiel’s fund offloaded Nvidia stake in third quarter, filing shows

0

Peter Thiel, pictured on the right [File]
| Photo Credit: REUTERS

Tech billionaire Peter Thiel’s hedge fund has sold off its entire stake in Nvidia during the third quarter, a regulatory filing showed, intensifying worries of an artificial intelligence bubble.

The fund, Thiel Macro, sold around 537,742 shares in the AI chip frontrunner in the quarter, the filing showed on Friday. The stake would have been worth around $100 million, as of the company’s closing price on September 30.

Thiel’s selloff, coupled with SoftBank’s sale of its own Nvidia holdings last week, has fueled Wall Street’s angst that the frenzy driving soaring tech valuations may have peaked, putting at risk the trillions of dollars committed to AI advancement.

Investors and analysts will be looking to Nvidia’s third-quarter results on Wednesday to dispel worries of a bubble as the world’s most valuable company is considered a bellwether for AI demand due to its coveted chips being used in massive data centers and servers.

Thiel Foundation did not immediately respond to a Reuters request for comment.

In the third quarter, several hedge funds trimmed their stakes in some of the largest seven tech firms, also called the “Magnificent Seven”, in a shift from their second-quarter activity, when leading stock-picking firms were more bullish on Big Tech names.

Thiel’s fund now counts Apple, Microsoft and a reduced stake in Tesla as its main holdings, according to the filing with the Securities and Exchange Commission.

Source link

X down for thousands of U.S. users, Downdetector shows

0

Image used for representational purposes.
| Photo Credit: Reuters

Elon Musk’s X was down for thousands of users in the U.S. on Tuesday (November 18, 2025), according to Downdetector.com.

There were more than 5,600 reports of issues with the social media platform, as of 6:51 a.m. ET, according to Downdetector, which tracks outages by collating status reports from a number of sources.

Cloudflare, a web infrastructure company, was also having issues that impacted other services. It was not immediately clear whether the outages were related.

Around 6:40 a.m. ET, Cloudflare said on its status page that it was investigating the issue. The company’s shares fell 4.1% in premarket trading.

“We are working to understand the full impact and mitigate this problem. More updates to follow shortly,” it said.

X and Cloudflare did not immediately respond to Reuters’ requests for comment.

Source link

Wall Street regulator drops emphasis on crypto sector exams for 2026

0

Under U.S. President Donald Trump, who has embraced the crypto sector politically and personally, the SEC has laid out a sweeping agenda to promote the development of the digital asset sector [File]
| Photo Credit: REUTERS

The U.S. Securities and Exchange Commission on Monday dropped its emphasis on the oversight of companies offering crypto asset-related services as part of its priorities for examining Wall Street firms for the current fiscal year, according to an annual statement published by the agency.

The SEC’s Division of Examinations, which scrutinises the legal compliance of investment advisers, broker-dealers, clearing agencies, stock exchanges and others, said it would focus on fiduciary duty, standards of conduct, and asset custody as well as new requirements for customer data privacy, among other subjects.

However, the statement contained no standalone section explicitly focusing on crypto activity and the volatility of digital assets, as it has in prior years. The U.S. government’s current fiscal year ends on September 30, 2026.

Under U.S. President Donald Trump, who has embraced the crypto sector politically and personally, the SEC has laid out a sweeping agenda to promote the development of the digital asset sector, marking an about-face from the prior administration, which viewed the industry as rife with fraud and noncompliance. The industry is likely to interpret Monday’s shift in emphasis as another encouraging sign.

In response to a request for comment, an SEC spokesperson referred to a passage in Monday’s announcement according to which this year’s priorities were “not … an exhaustive list of all areas” SEC examiners will focus on.

“Examinations are an important component to accomplishing the agency’s mission, but they should not be a ’gotcha’ exercise,” SEC Chairman Paul Atkins said in the announcement. “Today’s release of examination priorities should enable firms to prepare to have a constructive dialogue with SEC examiners and provide transparency into the priorities of the agency’s most public-facing division.”

Source link

Paul McCartney to release silent AI protest song

0

McCartney’s contribution is to the album “Is This What We Want” [File]
| Photo Credit: REUTERS

Pop legend Paul McCartney will release a silent music track next month as part of a silent album to protest UK copyright law changes that would give exemptions to tech firms.

Other artists such as Hans Zimmer and singer Kate Bush have joined the project, highlighting what they say are the dangers artificial intelligence (AI) poses to the creative industries.

McCartney’s contribution is to the album “Is This What We Want”. It will draw “attention to the damning impact on artists’ livelihoods controversial government proposals could cause,” the artists behind the project said in a statement.

Called “Bonus Track” it is a two minute 45 seconds recording of an empty studio featuring a series of clicks.

More than 1,000 artists, including Annie Lennox, Damon Albarn and Jamiroquai, have collaborated on the silent album which was first released in February.

They maintain that the government’s law changes “would make it easier to train AI models on copyrighted work without a licence”.

“Under the heavily criticised proposals, UK copyright law would be upended to benefit global tech giants. AI companies would be free to use an artist’s work to train their AI models without permission or remuneration,” they added.

The changes “would require artists to proactively ‘opt-out’ from the theft of their work – reversing the very principle of copyright law,” they added.

Only 1,000 copies of the vinyl album have been pressed.

In May, some 400 writers and musicians including Elton John and Bush condemned the proposals as a “wholesale giveaway” to Silicon Valley in a letter to The Times newspaper.

Other signatories included the 83-year-old McCartney, singer-songwriters Ed Sheeran, Dua Lipa and Sting, and writers Kazuo Ishiguro, Michael Morpurgo and Helen Fielding.

Prime Minister Keir Starmer has previously said the government needs to “get the balance right” with copyright and AI while noting that the technology represented “a huge opportunity”.

“They have no right to sell us down the river,” Elton John told the BBC in May, urging Starmer to “wise up” and “see sense”.

According to a study by UK Music last week two out of three artists and producers fear that AI poses a threat to their careers.

More than nine out of 10 surveyed demanded that their image and voice to be protected and called for AI firms to pay for the use of their creations.

Source link

Ciudades fracasan en reducir las muertes por accidentes de tránsito

0

LOS ÁNGELES, CA. — Kris Edwards esperaba en casa con unos amigos a que su esposa, Erika “Tilly” Edwards, regresara para salir a cenar; pero ella nunca volvió a la casa que habían comprado apenas cuatro días antes. Alrededor de las 9 p.m. del 29 de junio, un conductor que se dio a la fuga atropelló y mató a Tilly cuando caminaba hacia su auto tras asistir a un evento de recaudación de fondos en Hollywood.

“Tengo que encontrar la forma de seguir viviendo. Y lo más difícil es no saber por qué”, dijo Edwards sobre la muerte de su esposa.

A pesar de las campañas de seguridad vial impulsadas por autoridades locales, estatales y federales —como la iniciativa global Visión Cero que busca eliminar las muertes por accidentes de tránsito—, estas fatalidades han aumentado un 20% en Estados Unidos con respecto a hace una década: de 32.744 en 2014 a 39.345 en 2024, según datos de la Administración Nacional de Seguridad del Tráfico en las Carreteras del Departamento de Transporte (NHTSA).

Si bien las muertes han disminuido desde su punto más alto en 2021 (43.230), la cifra actual sigue siendo mayor que la de hace diez años.

Desde la pandemia de covid-19, el Pew Research Center ha identificado un deterioro en los hábitos de conducción de los estadounidenses, con aumentos en conductas como manejar de forma temeraria o bajo los efectos del alcohol, algo que defensores de la seguridad en las calles califican como un fracaso en salud pública.

Aunque aseguran que la tecnología podría reducir drásticamente estas muertes, las propuestas a menudo enfrentan resistencia de la industria, y la administración Trump está centrando sus esfuerzos en los autos sin conductor como forma de innovar y mejorar la seguridad pública.

“Cada día, 20 personas salen a caminar y no regresan a sus casas”, dijo Adam Snider, vocero de la Governors Highway Safety Association, que representa a las oficinas estatales de seguridad vial.

En algunas ciudades, las carreteras se han vuelto más peligrosas que los crímenes violentos. Los Ángeles, San Francisco y Houston son algunas de las grandes urbes que ahora reportan más muertes por accidentes de tránsito que por homicidios. En 2024, el Departamento de Policía de Los Ángeles registró unas 268 muertes por homicidio y 302 muertes por accidentes de tráfico, el segundo año consecutivo en que las víctimas de choques superan a las víctimas de homicidio, según Crosstown LA, un medio de comunicación comunitario sin fines de lucro.

San Francisco reportó 42 muertes por accidentes de tránsito y 35 homicidios en 2024. En Houston, aproximadamente 345 personas murieron en choques y 322 fueron víctimas de homicidio.

“En pocas palabras, Estados Unidos está en medio de una emergencia de seguridad vial”, declaró David Harkey, presidente del Instituto de Seguros para la Seguridad en las Carreteras (Insurance Institute for Highway Safety), durante una audiencia este verano en el subcomité de Energía y Comercio de la Cámara de Representantes. Harkey dijo que, entre 29 países de altos ingresos, Estados Unidos ocupa el último lugar en seguridad vial. “Este aumento no es —repito, no es— una tendencia global. Estados Unidos es una excepción”.

Kris Edwards y su gato, Rex, en el jardín de la casa que compró con su esposa, Erika “Tilly” Edwards, solo cuatro días antes de su muerte.(Chaseedaw Giles/KFF Health News)

En enero de 2017, el entonces alcalde Eric Garcetti, junto con otros 13 líderes de la ciudad, se comprometió a implementar el plan de acción Visión Cero y eliminar las muertes por accidentes de tránsito en Los Ángeles para 2025.

Sin embargo, las muertes han aumentado.

Una auditoría publicada en abril, encargada por la Oficina Administrativa de la Ciudad, concluyó que el entusiasmo por el programa se ha debilitado en el Ayuntamiento y que el proyecto ha sufrido debido a “la pandemia, conflictos personales, falta de compromiso total con su implementación, desacuerdos sobre cómo debía ser administrado y problemas para escalarlo”. El informe también señaló conflicto de intereses entre departamentos municipales e inversiones inconsistentes en los corredores viales más peligrosos de la ciudad.

La oficina de la alcaldesa Karen Bass no respondió de inmediato a las solicitudes de comentarios.

El año pasado, el senador estatal de California Scott Wiener propuso un proyecto de ley que habría exigido que todos los autos nuevos vendidos en el estado incluyeran un sistema de “asistencia inteligente de velocidad”, un software que evitaría que los vehículos superaran el límite de velocidad por más de 10 millas por hora.

Pero el proyecto fue debilitado tras la oposición de la industria automotriz y de algunos legisladores que lo consideraron una extralimitación del gobierno. Finalmente, fue vetado por el gobernador demócrata Gavin Newsom, quien argumentó que un mandato estatal interferiría con evaluaciones federales de seguridad que están en curso.

Mientras tanto, la Alianza para la Innovación Automotriz (Alliance for Automotive Innovation), un influyente grupo de presión de la industria automotriz, demandó este año al gobierno federal por una norma sobre frenos automáticos de emergencia implementada durante la administración de Joe Biden. La demanda está en curso en un tribunal federal mientras el Departamento de Transporte revisa la normativa. Incluso antes de que Donald Trump asumiera su segundo mandato, la alianza ya había enviado una carta al presidente electo solicitando su apoyo para defender la libertad de elección de los consumidores.

Durante la actual administración de Trump, el secretario de Transporte, Sean Duffy, ha dado prioridad al desarrollo de vehículos autónomos mediante propuestas de reformas regulatorias amplias para probar y lanzar autos sin conductor. “Las Normas Federales de Seguridad para Vehículos Motorizados fueron escritas para vehículos con conductores humanos y necesitan ser actualizadas para los vehículos autónomos”, dijo en septiembre Peter Simshauser, principal asesor legal de la NHTSA, al anunciar este esfuerzo de modernización, que incluye derogar algunas normas de seguridad. “Eliminar estos requisitos reducirá costos y mejorará la seguridad”.

Sin embargo, algunos legisladores demócratas han criticado la eliminación de normas de seguridad, calificándola como un error, ya que es posible implementar nuevas reglas sin deshacer protecciones existentes. Funcionarios de la NHTSA no respondieron a las solicitudes de comentarios sobre estas preocupaciones.

A photo of a husband and wife on a refrigerator magnet.
Una foto del compromiso de Kris Edwards y su esposa, Tilly, quien murió atropellada por un conductor que se dio a la fuga, en junio.(Chaseedaw Giles/KFF Health News)

Defensores de la seguridad vial temen que, si no se continúan adoptando regulaciones para los vehículos convencionales, factores como el exceso de velocidad y el error humano seguirán provocando muertes, a pesar del impulso hacia los autos sin conductor.

“Necesitamos seguir colaborando entre los sectores federal, estatal y local; el sector público y el privado; y también el público general”, señaló Snider, de la Asociación de Seguridad Vial de los Gobernadores. “Necesitamos que las personas que manejan se involucren”.

Pasó casi un mes antes de que la policía identificara al conductor de un Mercedes-Benz G-Wagen supuestamente implicado en la muerte de Tilly. Las autoridades acusaron a Davontay Robins de homicidio involuntario por negligencia grave en la conducción de un vehículo, de un delito grave por atropello con fuga y de manejar con una licencia suspendida debido a una condena previa por manejar alcoholizado. Robins se ha declarado inocente de todos los cargos y está en libertad bajo fianza.

Ahora, Kris Edwards cuida en soledad el jardín trasero de la casa que compartía con su esposa. Desde la muerte de Tilly, ha tenido problemas para dormir, fatiga, pérdida del apetito y necesita un bastón para caminar. Sus médicos atribuyen estos síntomas a la forma en que el cerebro responde al duelo.

“No estoy solo”, dijo. “Pero sí me siento solo, en esta casa grande y vacía sin mi compañera”.

Edwards espera que se haga justicia por la muerte de su esposa, aunque dice no estar seguro de que los fiscales logren una condena. Quiere que su muerte signifique algo: calles más seguras, conductores más lentos y peatones más cuidadosos al entrar y salir de sus autos en calles transitadas.

“Quiero que la muerte de mi esposa sea una advertencia para quienes se confían demasiado y bajan la guardia, aunque sea por un momento”, agregó. “Todo puede ocurrir en un instante”.



Source link

Google spurs renewable energy push for Indian Ocean data hub on Christmas Island

0

Australia’s infrastructure department is in discussions with Google to ensure its energy requirements are met without impacting supply to Christmas Island’s residents and businesses [File]
| Photo Credit: REUTERS

Australia’s remote Indian Ocean outpost of Christmas Island has enough power to support a new Google data centre without depriving locals, but its arrival could spur a push to renewable energy, the island’s biggest employer and the tech giant said.

Alphabet’s Google announced on Monday it will build a data hub on the tiny island located 350 km (220 miles) south of Indonesia, confirming a Reuters report.

It said it would also build a subsea cable system connecting Christmas Island to the Maldives and Oman, with two new data hubs to “deepen the resilience of internet infrastructure in the Indian Ocean Region”.

Google’s plans had raised concern there would not be enough power to meet the needs of locals, the island’s phosphate mine and the data centre, but Phosphate Resources chief executive, Nicholas Gan, said supply was ample for now.

The phosphate company, which employs half the island’s population of 1,600, imports diesel to run a power generator that supplies the mine and meets Australian defence force needs.

“The power grid can supply both Google’s requirements and our requirements comfortably,” Gan said.

Capacity would be strained, however, if the island’s detention centre for asylum seekers or a shuttered resort were to reopen, he said, adding that Google’s arrival bolsters the case for switching to renewable energy, which would be cheaper than importing diesel.

Australia’s infrastructure department is in discussions with Google to ensure its energy requirements are met without impacting supply to Christmas Island’s residents and businesses.

Another two planned Google subsea cables stretching east from Christmas Island will land near key Australian military bases, Reuters previously reported. Military experts say such a facility on the island would be valuable for using AI drones to monitor Chinese submarine activity.

Google said the island’s data hub would be smaller than some other Google data centres, and it would share its digital infrastructure with local users.

“The power required for a connectivity hub can still be a lot for some smaller locations, and where it is, Google is exploring using its power demand to accelerate local investment in sustainable energy generation,” its statement said.

A member of the island’s economic future working group, Gan said Google’s project will bring economic activity to an island 1,600 km from mainland Australia, with a history of boom and bust cycles, as it faced “the last era for mining”.

About 23 years ago, Australia and Russia planned to build a commercial spaceport on Christmas Island, but it prompted concern from Indonesia and was never opened, Australian parliament records show.

A casino that opened in 1993 attracted high rollers arriving by private jet from Jakarta but closed five years later amid an Asian economic downturn, the records show.

A detention centre for asylum seekers trying to reach Australia by boat dominated the island for two decades, until a shift in Australia’s immigration policy saw it largely emptied in 2023.

Source link

Ticked Off Over Preauthorization: Walk-In Patient Avoided Lyme Disease but Not a Surprise Bill

0

Leah Kovitch was pulling invasive plants in the meadow near her home one weekend in late April when a tick latched onto her leg.

She didn’t notice the tiny bug until Monday, when her calf muscle began to feel sore. She made an appointment that morning with a telehealth doctor — one recommended by her health insurance plan — who prescribed a 10-day course of doxycycline to prevent Lyme disease and strongly suggested she be seen in person. So, later that day, she went to a walk-in clinic near her home in Brunswick, Maine.

And it’s a good thing she did. Clinic staffers found another tick on her body during the same visit. Not only that, one of the ticks tested positive for Lyme, a bacterial infection that, if untreated, can cause serious conditions affecting the nervous system, heart, and joints. Clinicians prescribed a stronger, single dose of the prescription medication.

“I could have gotten really ill,” Kovitch said.

But Kovitch’s insurer denied coverage for the walk-in visit. Its reason? She hadn’t obtained a referral or preapproval for it. “Your plan doesn’t cover this type of care without it, so we denied this charge,” a document from her insurance company explained.

Health insurers have long argued that prior authorization — when health plans require approval from an insurer before someone receives treatment — reduces waste and fraud, as well as potential harm to patients. And while insurance denials are often associated with high-cost care, such as cancer treatment, Kovitch’s tiny tick bite exposes how prior authorization policies can apply to treatments that are considered inexpensive and medically necessary.

Kovitch and her partner often work in the garden at home and in an adjacent meadow. “We have chickens, so I’m just outside a lot,” she says. “In the springtime, we’re pulling ticks off us every day.”(Brianna Soukup for KFF Health News)

Pledging To Fix the Process

The Trump administration announced this summer that dozens of private health insurers agreed to make sweeping changes to the prior authorization process. The pledge includes releasing certain medical services from prior authorization requirements altogether. Insurers also agreed to extend a grace period to patients who switch health plans, so they won’t immediately encounter new preapproval rules that disrupt ongoing treatment.

Mehmet Oz, administrator of the Centers for Medicare & Medicaid Services, said during a June press conference that some of the changes would be in place by January. But, so far, the federal government has offered few specifics about which diagnostic codes tagged to medical services for billing purposes will be exempt from prior authorization — or how private companies will be held accountable. It’s not clear whether Lyme disease cases like Kovitch’s would be exempt from preauthorization.

Chris Bond, a spokesperson for AHIP, the health insurance industry’s main trade group, said that insurers have committed to implementing some changes by Jan. 1. Other parts of the pledge will take longer. For example, insurers agreed to answer 80% of prior authorization approvals in “real time,” but not until 2027.

Andrew Nixon, a spokesperson for the U.S. Department of Health and Human Services, told KFF Health News that the changes promised by private insurers are intended to “cut red tape, accelerate care decisions, and encourage transparency,” but they will “take time to achieve their full effect.”

Meanwhile, some health policy experts are skeptical that private insurers will make good on the pledge. This isn’t the first time major health insurers have vowed to reform prior authorization.

Bobby Mukkamala, president of the American Medical Association, wrote in July that the promises made by health insurers in June to fix the system are “nearly identical” to those the insurance industry put forth in 2018.

“I think this is a scam,” said Neal Shah, author of the book “Insured to Death: How Health Insurance Screws Over Americans — And How We Take It Back.”

Insurers signed on to President Donald Trump’s pledge to ease public pressure, Shah said. Collective outrage directed at insurance companies was particularly intense following the killing of UnitedHealthcare CEO Brian Thompson in December. Oz specifically said that the pledge by health insurers was made in response to “violence in the streets.”

Shah, for one, doesn’t believe companies will follow through in a meaningful way.

“The denials problem is getting worse,” said Shah, who co-founded Counterforce Health, a company that helps patients appeal insurance denials by using artificial intelligence. “There’s no accountability.”

Cracking the Case

A photo of Leah Kovitch showing her $238 bill.
After Kovitch sought care at a walk-in clinic for a tick bite, she learned her insurer would not cover the cost of the visit because it said she had not obtained a referral or preapproval. She tried appealing the insurer’s decision to no avail, eventually paying $238 out-of-pocket for the care she received at the clinic.(Brianna Soukup for KFF Health News)

Kovitch’s bill for her clinic appointment was $238, and she paid for it out-of-pocket after learning that her insurance company, Anthem, didn’t plan to cover a cent. First, she tried appealing the denial. She even obtained a retroactive referral from her primary care doctor supporting the necessity of the clinic visit.

It didn’t work. Anthem again denied coverage for the visit. When Kovitch called to learn why, she said she was left with the impression that the Anthem representative she spoke to couldn’t figure it out.

“It was like over their heads or something,” Kovitch said. “This was all they would say, over and over again: that it lacked prior authorization.”

Jim Turner, a spokesperson for Anthem, later attributed Kovitch’s denials to “a billing error” made by MaineHealth, the health system that operates the walk-in clinic where she sought care. MaineHealth’s error “resulted in the claim being processed as a specialist visit instead of a walk-in center/urgent care visit,” Turner told KFF Health News.

He did not provide documentation demonstrating how the billing error occurred. Medical records supplied by Kovitch show MaineHealth coded her walk-in visit as “tick bite of left lower leg, initial encounter,” and it’s not clear why Anthem interpreted that as a specialist visit.

After KFF Health News contacted Anthem with questions about Kovitch’s bill, Turner said that the company “should have identified the billing error sooner in the process than we did and we apologize for the confusion this caused Ms. Kovitch.”

Caroline Cornish, a spokesperson for MaineHealth, said this isn’t the only time Anthem has denied coverage for patients seeking walk-in or urgent care at MaineHealth. She said Anthem’s processing rules are sometimes misapplied to walk-in visits, leading to “inappropriate denials.”

She said these visits should not require prior authorization and Kovitch’s case illustrates how insurance companies often use administrative denials as a first response.

“MaineHealth believes insurers should focus on paying for the care their members need, rather than creating obstacles that delay coverage and risk discouraging patients from seeking care,” she said. “The system is too often tilted against the very people it is meant to serve.”

Meanwhile, in October, Anthem sent Kovitch an updated explanation of benefits showing that a combination of insurance company payments and discounts would cover the entire cost of the appointment. She said a company representative called her and apologized. In early November, she received her $238 refund.

But she recently found out that her annual eye appointment now requires a referral from her primary care provider, according to new rules laid out by Anthem.

“The trend continues,” she said. “Now I am more savvy to their ways.”

A photo of Leah Kovitch walking to her home from her meadow.
After KFF Health News approached Anthem with questions for this article, Kovitch’s insurer apologized and said she owed nothing for the clinic visit.(Brianna Soukup for KFF Health News)



Source link