×
Lisbon Glória funicular: What we know about deadly crash

Lisbon Glória funicular: What we know about deadly crash

Emily Atkinson, Malu Cursino & Patrick JacksonBBC News

Watch: Emergency crews surround derailed Portugal funicular

A carriage on one of Lisbon’s most iconic tourist attractions, the Glória funicular railway derailed and crashed on Wednesday. The popular tram-like form of transport is designed to travel up and down steep slopes.

Portuguese authorities have confirmed 16 people were killed, revising an earlier figure which put the death toll at 17.

It is not yet clear what caused the carriage to derail, but local media reports say a cable fault is suspected.

An investigation into the cause of the accident is under way.

Here is everything we know so far.

A satellite map of central Lisbon shows the location of where Lisbon's Gloria funicular crashed. Labels show the crash happened to the right of Rossio train station, close to Restauradores square and Avenida da Liberdade. An inset shows the hill in which the crash happened, showing a steep gradient.

What happened?

The carriage derailed at around 18:15 (17:15 GMT) on Wednesday 3 September, near the city’s Avenida da Liberdade boulevard.

More than 60 rescue personnel and 22 vehicles were deployed to the scene.

Officials say it is too early to determine what caused the crash, but Portuguese newspaper Observador reports that a cable came loose along the railway’s route, causing it to lose control and collide with a building.

As smoke engulfed the area, people could be seen running away from the scene.

Images and footage show an overturned, crumpled yellow carriage lying on the cobblestone street, surrounded by emergency responders.

Some people were trapped in the wreckage, and had to be freed by emergency workers.

Tour guide Marianna Figueiredo witnessed the crash and was amongst those who tried to rescue people.

“People started to jump from the windows inside the funicular at the bottom of the hill,” she told the BBC.

“Then I saw another one [further up] that was already crushed. I started to climb the hill to help the people but when I got there the only thing I could hear was silence.”

She said what she witnessed on Wednesday evening was “very difficult to describe”. “It was very bad. A big tragedy.”

Who are the casualties?

Emergency services gave a death toll of 17 on Thursday but Prime Minister Luís Montenegro later said 16 were dead.

Reports in Portugal have attributed the mistake in the death toll to a duplicate registration of a victim at one hospital.

Among those who died were five Portuguese, two South Koreans, two Canadians, an American, a German, a Ukrainian and a Swiss. Three victims remain unidentified.

Exact information on who was riding the funicular at the time – and who may have been injured on the street – remains unclear but a further 23 people have been injured according to Alvaro Santos Almeida, head of Portugal’s’ health service.

Six of those are in intensive care while three sustained minor wounds.

Three of those injured are Portuguese, one is German, one is South Korean, one is Swiss, one is Cape Verdean and one is Moroccan.

André Jorge Gonçalves Marques, an employee of the city’s public transport operator, Carris, is among the dead.

“In his 15 years with Carris, he performed his duties with excellency,” the company said.

Portuguese media is reporting that a three-year-old German child was rescued from the wreckage, while the child’s father was killed and the mother injured.

The Glória funicular can carry about 40 passengers and is extremely popular with tourists but it is also crucial for the city’s residents, to help them travel up and down Lisbon’s hilly streets.

It is not known how many people were onboard at the time of the crash, however.

What is the Glória funicular and how does it work?

Two yellow funicular carriages on a steep 18% incline in Lisbon, connected by an underground cable. Overhead power lines and tracks are labelled. The brakeman is visible in the carriage on the left, which is descending. A person walks uphill beside the carriage on the right.

A funicular is a type of railway system that allows travel up and down steep slopes, and in Lisbon, they are a crucial means of navigating the city’s steep, cobbled streets.

The city’s funicular railways – Glória, Lavra, Bica and Graça – are a popular tourist attraction, as the bright yellow tram-like vehicles snake through the often-narrow, hilly streets.

Glória was opened in 1885 and electrified three decades later.

It travels some 275m (900ft) from Restauradores, a central city square, up to the picturesque streets of Bairro Alto. The journey takes just three minutes.

The two carriages on the Glória route are attached to opposite ends of a haulage cable which is pulled by electric motors.

As one carriage travels downhill, its weight lifts the other, allowing them to ascend and descend simultaneously, reducing the energy need.

The second, intact carriage could be seen just metres from the wreckage at the bottom of the hill.

Reuters An intact yellow and white tram stands a short distance from the mangled wreckage of another tram on a narrow, sloping city street.Reuters

How safe are Lisbon’s funicular railways?

Expressing regret for the crash, Carris said it had opened an investigation “to determine the causes of this accident”.

The company said in a statement that it had complied with “all maintenance protocols”:

  • General maintenance took place every four years and was last carried out in 2022
  • Interim repairs were made every two years, the last of which took place in 2024
  • Monthly and weekly maintenance programmes and daily inspections were also “scrupulously complied with”

“Everything was scrupulously respected,” company head Pedro Bogas said separately, adding that maintenance of the funiculars had been carried out by a contractor for the past 14 years.

“We have strict protocols, excellent professionals for many years, and we need to get to the bottom of what happened,” he added.

How have people reacted to the crash?

The country was observing a national day of mourning on Thursday while Lisbon Mayor Carlos Moedas declared three days of mourning in the capital.

Posting on X, Moedas said: “I extend my heartfelt condolences to all the families and friends of the victims. Lisbon is in mourning.”

Portuguese President Marcelo Rebelo de Sousa said he “deeply” regretted the “fatalities and serious injuries” caused by the crash.

Pedro Sánchez, prime minster of neighbouring Spain, said he was “appalled by the terrible accident”.

“All our affection and solidarity with the families of the victims and with the Portuguese people in this difficult moment,” he wrote in a statement on X.

European Commission President Ursula von der Leyen sent her “condolences to the families of the victims”.

In an interview with BBC Radio 4’s Today programme, Fabiana Pavel, who is president of a Lisbon residents’ association, said her community was shocked.

She said the funicular is an important means of public transportation for those who live in Bairro Alto, “especially for people with reduced mobility as it allows them to easily climb a steep hill”.

“At the time of the accident there may have been parents with children returning from school.”

Ms Pavel added that the cable car is used disproportionately by tourists, leaving locals unable to use the service as a means of transport “because it has become a tourist attraction”.

Flags outside the European Parliament in Brussels have been flying at half mast to mark Portugal’s day of mourning.

“The tragic accident on the Elevador da Gloria has shaken Europe deeply,” Roberta Metsola, president of the European Parliament, posted on X.

Source link
#Lisbon #Glória #funicular #deadly #crash

岩手 大槌町 山林火災は住宅近く迫る 1000人規模で消火 | NHKニュース岩手県大槌町の山林火災は25日で発生から4日目となりますが、延焼が続いています。町によりますと複数の地区にある住宅の近くまで火が迫っているということで、1000人以上の規模での懸命な消火活動が行われています。#岩手 #大槌町 #山林火災は住宅近く迫る #1000人規模で消火 #NHKニュースNHK,ニュース,NHK ONE,火災,岩手県,一覧

OpenAI CEO Sam Altman has apologized to the Canadian town of Tumbler Ridge following a February mass shooting that left eight dead. 

Altman said he was “deeply sorry” the company didn’t alert the police about the shooter’s troubling ChatGPT accounts.

Britich Colombia Primier David Eby called the apology “necessary, and yet grossly insufficient.”

How did OpenAI fail to act?

An 18-year-old transgender woman killed her mother and stepbrother at home on February 10, before going to a local secondary school and opening fire. She killed five children and a teacher, then took her own life.

After the attack, OpenAI said it had identified the suspect’s account through its abuse detection systems and banned it in June, eight months before the shooting.

The ChatGPT developer said it did not report the account to Canadian police at the time, as the activity did not meet its threshold for referral to law enforcement.

“I am deeply sorry that we did not alert law enforcement to the account that was banned in June,” Altman said. “While I know words can never be enough, I believe an apology is necessary to recognize the harm and irreversible loss your community has suffered.” 

How does ChatGPT report suspected violance?

OpenAI says it uses automated moderation systems that scan content in real time. Accounts can be restricted or banned for violating the rules. Violations include sexual exploitation, support of self-harm and suicide, and promotion of violence and harm.

In serious cases, systems are designed to flag high-risk behavior for human review. If a credible threat is identified, the company may share relevant account data with law enforcement.

Following the attack, Canadian officials summoned OpenAI’s safety team and warned of regulation actions if changes were not made. The company said it would tighten its safety measures and had created a direct contact channel with police.

In the letter, Altman said the company is committed to find ways to prevent similar tragedies. “Going forward, our focus will continue to be on working with all levels of government to help ensure something like this never happens again,” he said. 

The family of a girl who was seriously injured in the shooting has filed a negligence lawsuit against the US tech giant.

Is your AI private? OpenAI and the Canadian school shooting

Edited by: Wesley Dockery 

#OpenAI #apologizes #reporting #Canada #mass #shooter">OpenAI apologizes for not reporting Canada mass shooterOpenAI CEO Sam Altman has apologized to the Canadian town of Tumbler Ridge following a February mass shooting that left eight dead. 

Altman said he was “deeply sorry” the company didn’t alert the police about the shooter’s troubling ChatGPT accounts.

Britich Colombia Primier David Eby called the apology “necessary, and yet grossly insufficient.”

How did OpenAI fail to act?

An 18-year-old transgender woman killed her mother and stepbrother at home on February 10, before going to a local secondary school and opening fire. She killed five children and a teacher, then took her own life.

After the attack, OpenAI said it had identified the suspect’s account through its abuse detection systems and banned it in June, eight months before the shooting.

The ChatGPT developer said it did not report the account to Canadian police at the time, as the activity did not meet its threshold for referral to law enforcement.

“I am deeply sorry that we did not alert law enforcement to the account that was banned in June,” Altman said. “While I know words can never be enough, I believe an apology is necessary to recognize the harm and irreversible loss your community has suffered.” 

How does ChatGPT report suspected violance?

OpenAI says it uses automated moderation systems that scan content in real time. Accounts can be restricted or banned for violating the rules. Violations include sexual exploitation, support of self-harm and suicide, and promotion of violence and harm.

In serious cases, systems are designed to flag high-risk behavior for human review. If a credible threat is identified, the company may share relevant account data with law enforcement.

Following the attack, Canadian officials summoned OpenAI’s safety team and warned of regulation actions if changes were not made. The company said it would tighten its safety measures and had created a direct contact channel with police.

In the letter, Altman said the company is committed to find ways to prevent similar tragedies. “Going forward, our focus will continue to be on working with all levels of government to help ensure something like this never happens again,” he said. 

The family of a girl who was seriously injured in the shooting has filed a negligence lawsuit against the US tech giant.

Is your AI private? OpenAI and the Canadian school shootingTo view this video please enable JavaScript, and consider upgrading to a web browser that supports HTML5 video

Edited by: Wesley Dockery 
#OpenAI #apologizes #reporting #Canada #mass #shooter

February mass shooting that left eight dead. 

Altman said he was “deeply sorry” the company didn’t alert the police about the shooter’s troubling ChatGPT accounts.

Britich Colombia Primier David Eby called the apology “necessary, and yet grossly insufficient.”

How did OpenAI fail to act?

An 18-year-old transgender woman killed her mother and stepbrother at home on February 10, before going to a local secondary school and opening fire. She killed five children and a teacher, then took her own life.

After the attack, OpenAI said it had identified the suspect’s account through its abuse detection systems and banned it in June, eight months before the shooting.

The ChatGPT developer said it did not report the account to Canadian police at the time, as the activity did not meet its threshold for referral to law enforcement.

“I am deeply sorry that we did not alert law enforcement to the account that was banned in June,” Altman said. “While I know words can never be enough, I believe an apology is necessary to recognize the harm and irreversible loss your community has suffered.” 

How does ChatGPT report suspected violance?

OpenAI says it uses automated moderation systems that scan content in real time. Accounts can be restricted or banned for violating the rules. Violations include sexual exploitation, support of self-harm and suicide, and promotion of violence and harm.

In serious cases, systems are designed to flag high-risk behavior for human review. If a credible threat is identified, the company may share relevant account data with law enforcement.

Following the attack, Canadian officials summoned OpenAI’s safety team and warned of regulation actions if changes were not made. The company said it would tighten its safety measures and had created a direct contact channel with police.

In the letter, Altman said the company is committed to find ways to prevent similar tragedies. “Going forward, our focus will continue to be on working with all levels of government to help ensure something like this never happens again,” he said. 

The family of a girl who was seriously injured in the shooting has filed a negligence lawsuit against the US tech giant.

Is your AI private? OpenAI and the Canadian school shooting

Edited by: Wesley Dockery 

#OpenAI #apologizes #reporting #Canada #mass #shooter">OpenAI apologizes for not reporting Canada mass shooter

OpenAI CEO Sam Altman has apologized to the Canadian town of Tumbler Ridge following a February mass shooting that left eight dead. 

Altman said he was “deeply sorry” the company didn’t alert the police about the shooter’s troubling ChatGPT accounts.

Britich Colombia Primier David Eby called the apology “necessary, and yet grossly insufficient.”

How did OpenAI fail to act?

An 18-year-old transgender woman killed her mother and stepbrother at home on February 10, before going to a local secondary school and opening fire. She killed five children and a teacher, then took her own life.

After the attack, OpenAI said it had identified the suspect’s account through its abuse detection systems and banned it in June, eight months before the shooting.

The ChatGPT developer said it did not report the account to Canadian police at the time, as the activity did not meet its threshold for referral to law enforcement.

“I am deeply sorry that we did not alert law enforcement to the account that was banned in June,” Altman said. “While I know words can never be enough, I believe an apology is necessary to recognize the harm and irreversible loss your community has suffered.” 

How does ChatGPT report suspected violance?

OpenAI says it uses automated moderation systems that scan content in real time. Accounts can be restricted or banned for violating the rules. Violations include sexual exploitation, support of self-harm and suicide, and promotion of violence and harm.

In serious cases, systems are designed to flag high-risk behavior for human review. If a credible threat is identified, the company may share relevant account data with law enforcement.

Following the attack, Canadian officials summoned OpenAI’s safety team and warned of regulation actions if changes were not made. The company said it would tighten its safety measures and had created a direct contact channel with police.

In the letter, Altman said the company is committed to find ways to prevent similar tragedies. “Going forward, our focus will continue to be on working with all levels of government to help ensure something like this never happens again,” he said. 

The family of a girl who was seriously injured in the shooting has filed a negligence lawsuit against the US tech giant.

Is your AI private? OpenAI and the Canadian school shooting

Edited by: Wesley Dockery 

#OpenAI #apologizes #reporting #Canada #mass #shooter

Post Comment