Thursday, November 04, 2021

Doomsday Clock now 100 seconds from midnight

clock
Credit: Pixabay/CC0 Public Domain

While the Doomsday Clock is perilously close to midnight, it is not as close as UK Prime Minister Boris Johnson recently suggested in his COP26 opening remarks

"Humanity has long since run down the clock on climate change," Johnson said. "It's one minute to midnight on that  and we need to act now."

The Bulletin of the Atomic Scientists is in complete agreement with the sentiment that "we need to act now," but would like to clarify that the Doomsday Clock, which it created in 1947, is currently set at 100 seconds to midnight.

The Clock, a powerful symbol for how close humanity is to self-annihilation, is set by the Bulletin's Science and Security Board once a year. The  weigh dangers posed by climate change, nuclear risk and disruptive technologies in determining the time. The board will announce the time in January 2022 and commemorate the Clock's 75th anniversary with an upcoming book.

As one of the most recognizable symbols in the past 100 years, the Doomsday Clock sits at the crossroads of science and art. It has permeated not only the media landscape, but culture itself. The anniversary book chronicles the Doomsday Clock's references in novels by writers such as Stephen King and Piers Anthony, comic books (Watchmen, StormWatch), movies (Justice League), music (The Who, The Clash, Smashing Pumpkins) and numerous art exhibitions.

'Doomsday Clock' stays at two mins to midnight
More information: thebulletin.org/doomsday-clock/current-time/
This is your COVID wake-up call:
It is 100 seconds to midnight

2021 Doomsday Clock Statement

Science and Security Board
Bulletin of the Atomic Scientists
Editor, John Mecklin



It is 100 seconds to midnight



PDF / print version

Editor’s note: Founded in 1945 by Albert Einstein and University of Chicago scientists who helped develop the first atomic weapons in the Manhattan Project, the Bulletin of the Atomic Scientists created the Doomsday Clock two years later, using the imagery of apocalypse (midnight) and the contemporary idiom of nuclear explosion (countdown to zero) to convey threats to humanity and the planet. The Doomsday Clock is set every year by the Bulletin’s Science and Security Board in consultation with its Board of Sponsors, which includes 13 Nobel laureates. The Clock has become a universally recognized indicator of the world’s vulnerability to catastrophe from nuclear weapons, climate change, and disruptive technologies in other domains.


To: Leaders and citizens of the world

Re: This is your COVID wake-up call: It is 100 seconds to midnight

Date: January 27, 2021

Humanity continues to suffer as the COVID-19 pandemic spreads around the world. In 2020 alone, this novel disease killed 1.7 million people and sickened at least 70 million more. The pandemic revealed just how unprepared and unwilling countries and the international system are to handle global emergencies properly. In this time of genuine crisis, governments too often abdicated responsibility, ignored scientific advice, did not cooperate or communicate effectively, and consequently failed to protect the health and welfare of their citizens.

As a result, many hundreds of thousands of human beings died needlessly.

Though lethal on a massive scale, this particular pandemic is not an existential threat. Its consequences are grave and will be lasting. But COVID-19 will not obliterate civilization, and we expect the disease to recede eventually. Still, the pandemic serves as a historic wake-up call, a vivid illustration that national governments and international organizations are unprepared to manage nuclear weapons and climate change, which currently pose existential threats to humanity, or the other dangers—including more virulent pandemics and next-generation warfare—that could threaten civilization in the near future.

Accelerating nuclear programs in multiple countries moved the world into less stable and manageable territory last year. Development of hypersonic glide vehicles, ballistic missile defenses, and weapons-delivery systems that can flexibly use conventional or nuclear warheads may raise the probability of miscalculation in times of tension. Events like the deadly assault earlier this month on the US Capitol renewed legitimate concerns about national leaders who have sole control of the use of nuclear weapons. Nuclear nations, however, have ignored or undermined practical and available diplomatic and security tools for managing nuclear risks. By our estimation, the potential for the world to stumble into nuclear war—an ever-present danger over the last 75 years—increased in 2020. An extremely dangerous global failure to address existential threats—what we called “the new abnormal” in 2019—tightened its grip in the nuclear realm in the past year, increasing the likelihood of catastrophe.

Governments have also failed to sufficiently address climate change. A pandemic-related economic slowdown temporarily reduced the carbon dioxide emissions that cause global warming. But over the coming decade fossil fuel use needs to decline precipitously if the worst effects of climate change are to be avoided. Instead, fossil fuel development and production are projected to increase. Atmospheric greenhouse gas concentrations hit a record high in 2020, one of the two warmest years on record. The massive wildfires and catastrophic cyclones of 2020 are illustrations of the major devastation that will only increase if governments do not significantly and quickly amplify their efforts to bring greenhouse gas emissions essentially to zero.

As we noted in our last Doomsday Clock statement, the existential threats of nuclear weapons and climate change have intensified in recent years because of a threat multiplier: the continuing corruption of the information ecosphere on which democracy and public decision-making depend. Here, again, the COVID-19 pandemic is a wake-up call. False and misleading information disseminated over the internet—including misrepresentation of COVID-19’s seriousness, promotion of false cures, and politicization of low-cost protective measures such as face masks—created social chaos in many countries and led to unnecessary death. This wanton disregard for science and the large-scale embrace of conspiratorial nonsense—often driven by political figures and partisan media—undermined the ability of responsible national and global leaders to protect the security of their citizens. False conspiracy theories about a “stolen” presidential election led to rioting that resulted in the death of five people and the first hostile occupation of the US Capitol since 1814.

In 2020, online lying literally killed.

Considered by themselves, these negative events in the nuclear, climate change, and disinformation arenas might justify moving the clock closer to midnight. But amid the gloom, we see some positive developments. The election of a US president who acknowledges climate change as a profound threat and supports international cooperation and science-based policy puts the world on a better footing to address global problems. For example, the United States has already announced it is rejoining the Paris Agreement on climate change and the Biden administration has offered to extend the New START arms control agreement with Russia for five years. In the context of a post-pandemic return to relative stability, more such demonstrations of renewed interest in and respect for science and multilateral cooperation could create the basis for a safer and saner world.

Because these developments have not yet yielded substantive progress toward a safer world, they are not sufficient to move the Clock away from midnight. But they are positive and do weigh against the profound dangers of institutional decay, science denialism, aggressive nuclear postures, and disinformation campaigns discussed in our 2020 statement. The members of the Science and Security Board therefore set the Doomsday Clock at 100 seconds to midnight, the closest it has ever been to civilization-ending apocalypse and the same time we set in 2020. It is deeply unfortunate that the global response to the pandemic over the past year has explicitly validated many of the concerns we have voiced for decades.

We continue to believe that human beings can manage the dangers posed by modern technology, even in times of crisis. But if humanity is to avoid an existential catastrophe—one that would dwarf anything it has yet seen—national leaders must do a far better job of countering disinformation, heeding science, and cooperating to diminish global risks. Citizens around the world can and should organize and demand—through public protests, at ballot boxes, and in other creative ways—that their governments reorder their priorities and cooperate domestically and internationally to reduce the risk of nuclear war, climate change, and other global disasters, including pandemic disease.

We have experienced the consequences of inaction. It is time to respond.



A dark nuclear landscape, with glimmers of hope


In the past year, countries with nuclear weapons continued to spend vast sums on nuclear modernization programs, even as they allowed proven risk-reduction achievements in arms control and diplomacy to wither or die. Nuclear weapons and weapons-delivery platforms capable of carrying either nuclear or conventional warheads continued to proliferate, while destabilizing “advances” in the space and cyber realms, in hypersonic missiles, and in missile defenses continued. Governments in the United States, Russia, and other countries appear to consider nuclear weapons more-and-more usable, increasing the risks of their actual use. There continues to be an extraordinary disregard for the potential of an accidental nuclear war, even as well-documented examples of frighteningly close calls have emerged.

US and Russian nuclear modernization efforts continued to accelerate, and North Korea, China, India, and Pakistan pursued “improved” and larger nuclear forces. Some of these modernization programs are beginning to field weapons with dangerous enhancements, like Russia’s nuclear-tipped Avangard hypersonic glide vehicles, which are being installed on new SS-29 (Sarmat) missiles designed to replace 1980s-era intercontinental ballistic missiles (ICBMs). Russia continues to field battalions of intermediate-range, ground-launched, nuclear-armed missiles—missiles previously banned by the now-defunct Intermediate-range Nuclear Forces Treaty, from which the United States withdrew in 2019. China, which has historically relied on a small and constrained nuclear arsenal, is expanding its capabilities and deploying multiple, independently retargetable warheads on some of its ICBMs and will likely add more in the coming year.



In December 2018, Russia’s Strategic Rocket Forces tested the Avangard system, which included a hypersonic glide vehicle carried on a UR-100NUTTH/SS-19 missile. (Russian Defense Ministry video via russianforces.org)

The heightened interest that the United States and Russia have shown in hypersonic weapons, as demonstrated by a number of tests in 2020, is deeply worrisome. The hypersonics arms race has already led to calls for space-based interceptors to destroy them in flight. This militarization of space is dangerously destabilizing and increases the risk of escalation and accidental conflict.

Several countries are developing weapons-delivery platforms that can carry either nuclear or conventional warheads, introducing greater risks of miscalculation in a crisis or conventional conflict. Some may view this ambiguity as a deterrent to war, but it is not hard to imagine how mistaking a conventionally armed cruise missile for a nuclear-armed missile could complicate decision-making in the fog of crisis or war, potentially leading to preemptive strikes. The potential to stumble into nuclear war—ever present—has grown.

Meanwhile, developments in Northeast Asia, the Middle East, and South Asia further add to nuclear risks.

North Korea continues to develop its missile and nuclear programs. It revealed a new and larger long-range missile (Hwasong-16) in October 2020 at a military parade, but in the absence of flight testing, it’s not clear whether the new missile will add major capabilities to North Korea’s arsenal. There were no high level meetings between North Korea and the United States in 2020, leaving the future of US negotiations with North Korea in doubt.


The Hwasong-16 in the parade of the 75th anniversary of the Worker’s Party of Korea in October 2020. (Photo credit: Wikimedia Commons)

South Asia remains a potential nuclear hot spot, as both India and Pakistan continue to enlarge their arsenals and increase the sophistication and ranges of their weapons, with Indian ballistic missiles now able to reach Chinese targets. The relatively recent movement of nuclear competition among these countries to sea-based platforms, including submarines, raises the risk—already high—that conventional skirmishes could escalate to the nuclear level.

The continued effort by Iran to enhance its nuclear capabilities is another serious concern. But a bright spot in an otherwise gloomy landscape is the Biden administration’s stated desire to rejoin the Iran nuclear deal, known officially as the Joint Comprehensive Plan of Action (JCPOA). In response to the 2018 US withdrawal, Iran deliberately walked back its commitments under the agreement. Stockpiles of low-enriched uranium have increased, enrichment levels have risen, and new, improved centrifuges have been installed. These actions have reduced the amount of time it would take Iran to put together a nuclear weapon from one year to several months. At the same time, Iran continues to comply with many of the agreement’s requirements, and many of the actions it has taken can easily be reversed. However, Iran’s willingness to remain in the agreement is not a given.

To keep nuclear modernization programs from becoming a full-scale nuclear arms race, it will be essential that New START, a treaty that limits US and Russian strategic weapon deployments, be extended for five more years, buying time for a follow-on agreement to be considered, negotiated, and put into force. Russian President Vladimir Putin and new US President Joe Biden agreed to do that on January 26 and now the action is in the Duma’s hands.

Other arms control efforts—including the nuclear test ban treaty and negotiations to stop producing fissile materials for weapons—have unraveled or are stalled. Previous cooperation on fissile material control and nuclear proliferation among the United States, Russia, and China has lapsed, and there are no serious efforts aimed at limiting risky developments in cyberweapons, space weapons, missile defenses, and hypersonic missiles.

The tenth review of the Non-Proliferation Treaty (NPT) was postponed in 2020 because of the COVID-19 pandemic. Rescheduled for this year, the review conference will provide an opportunity for nuclear weapons countries to demonstrate the practical steps they have taken or will commit to take to reduce the risks of nuclear weapons use and scale back their reliance on nuclear weapons.

Just a few days ago, the Treaty on the Prohibition of Nuclear Weapons entered into force after 50 countries completed ratification. This treaty was developed by countries that do not have nuclear weapons, with the intention of bringing pressure on the nuclear weapons states to move more forcefully toward nuclear disarmament. The treaty brings much-needed attention to the risks posed by nuclear weapons, especially the enormous humanitarian impacts of the use of nuclear weapons. We hope that the treaty will lead to concrete actions by all states to address the challenges of disarmament and proliferation, including collective security and verification. We call on all states to collaborate and compromise to achieve real disarmament results.


Climate change action after the pandemic

Last year was to have marked a climate change milestone: The parties to the Paris Agreement were expected to increase their pledges to reduce the greenhouse gas emissions that are disrupting Earth’s climate. The initial pledges made in 2015 to reduce emissions over this decade were markedly inadequate and meant only to begin an iterative process towards the goal of limiting global warming to well below 2 degrees Celsius, relative to pre-industrial levels. Countries had been expected to raise their pledges at the 2020 meeting, but because of the coronavirus pandemic, the meeting was postponed until this year.

The delay may help. Few countries have been paying much attention to climate action during the pandemic. In 2020, countries whose emissions amounted to barely one-quarter of the global total had submitted improved emissions pledges, and countries responsible for another quarter of global emissions—including Australia, Japan, the United States, Russia, Indonesia, Brazil and New Zealand—simply announced pledges that were effectively identical to or even weaker than their existing commitments. Although the United States formally withdrew from the Paris Agreement late last year, the new administration has begun the process of rejoining and expressed its intention to submit an improved pledge and to provide additional financial support for climate actions in poor countries. As the pandemic recedes, more countries may step up their pledges over the course of the coming year.

As the COVID-19 pandemic deepened in the early months of 2020, carbon dioxide emissions dropped by an estimated 17 percent compared to the previous year’s. Emissions have largely bounced back, however, as the world’s fossil fuel-dependent economies have begun to recover, and the year’s total emissions were estimated at only four-to-seven percent lower than last year’s. Of course, cutting emissions temporarily via disease-induced economic recession is neither desirable nor sustainable. And, as with other economic crises, further recovery will raise energy demand and thus emissions—unless we take deliberate policy steps to reduce fossil-fuel use and accelerate the adoption of alternatives.

Fortunately, renewable energy has been resilient in the turbulent pandemic energy environment. Renewable deployment has slowed, but by less than other sources, and investment remains high. In the US, coal is projected to provide less electricity than renewables for the first time ever, owing to a decline in electricity demand and coal’s inability to compete given the low price of natural gas and near-zero operating costs of renewables. Globally, demand for fossil-based power has declined, while demand for renewable power has risen.

These developments need to be sustained into the recovery from the COVID-19 crisis, but are not nearly enough to halt warming. Global greenhouse gas concentrations in the atmosphere have hit a record high, and 2020 was essentially tied with 2016 as the warmest year on record. Until global carbon dioxide emissions are reduced nearly to zero, the burden of carbon dioxide in the atmosphere will continue to mount, and the world will continue to warm. The climate is still heading in the wrong direction.

In 2020, the impacts of continuing climate change were underscored in extreme and damaging ways. Portions of North America and Australia suffered massive wildfires, and a clear signal of human-caused climate change was evident in the frequency of powerful tropical cyclones and the heavier rainfall they produced. Meanwhile, evidence mounted that sea level rise is accelerating, and the effects of the oceans growing warmer and more acidic because of carbon dioxide absorption were clear in many marine ecosystems, as was most dramatically illustrated by the ongoing destruction of coral reefs.


The Orroral Valley Fire viewed from Tuggeranon, a suburb of Canberra, Australia, on the evening of January 28, 2020. (Photo credit: Nick-D / Wikimedia Commons)

In the long term, the answers to two questions related to the pandemic will have important climate change ramifications:

First, to what extent will economic stimulus spending aimed at ending the coronavirus economic slowdown be directed toward efficient green infrastructure and low-carbon industries? Such support will inevitably compete with aid requests from fossil fuel companies and other carbon-intensive industries that are also facing pandemic-related pressures

In the COVID-19 case, a lot of “brown” (fossil-based) stimulus is in the works. The trillions of dollars in stimulus programs that countries have launched are not particularly green. In aggregate, the G20 countries had committed approximately $240 billion to stimulus spending that supports fossil fuel energy by the end of 2020, versus $160 billion for clean energy. Likewise, the support packages for developing countries from the World Bank and International Monetary Fund do not favor low-carbon investments. And while China has made strong commitments to the decarbonization of its domestic economy, its Belt and Road Initiative appears poised to fill the niche increasingly being abandoned by developed country finance sectors, pouring investment into fossil-fuel infrastructure around the world.

At present, national plans for fossil fuel development and production are anything but encouraging; they project global growth in carbon dioxide emissions from fossil fuel use of roughly two percent per year over this coming critical decade, whereas emissions would need to decline precipitously if the temperature commitments of the Paris Agreement were to be met. If these plans are indeed pursued, fossil fuel production in 2030 would be around 50 percent higher than is consistent with meeting even the least ambitious goals of the Paris Agreement.

A second question: How will the pandemic affect the ability of the international political system to manage global climate change? Like climate change, the COVID-19 pandemic is a global problem that calls for a global solution. How successfully the leaders of the world’s nations coordinate their responses to the pandemic affects (or, will affect) their faith and commitment to multilateralism generally. They could become more confident in the value of effective global cooperation and robust international institutions, or they could emerge more mistrustful of multilateralism and discard their remaining commitments to invest in already declining and over-stretched institutions of global cooperation. A positive experience could lead to effective collaborations addressing climate change, the threat of nuclear war, and global challenges yet to emerge.


The COVID-19 infodemic and other disruptive threats


The COVID-19 pandemic has disrupted the planet in many extraordinary and negative ways, one of which involves the internet-driven spread of false or misleading information. As the pandemic emerged, it spawned what the World Health Organization has called a “massive ‘infodemic’—an over-abundance of information … that makes it hard for people to find trustworthy sources and reliable guidance when they need it.” The COVID-19 infodemic includes deliberate attempts (sometimes by national leaders) to disseminate misinformation and disinformation that harms physical and mental health; threatens public health gains; damages economies; and makes it much more difficult for the nations of the world to stop the pandemic.

The COVID-19 pandemic and its accompanying infodemic have become intertwined with critical uncertainties regarding science, technology, and crisis communications.

First, not all of the science relevant to ending the pandemic was known at its outset. Alas, many loud voices regarded the evolution of scientific knowledge about COVID-19 as reason to ignore and disparage scientific advice about controlling the pandemic.

Also, as new science-based treatments and interventions were developed and tested, experts needed to learn how to maximize their beneficial effects and deliver them to the public. This learning process introduced uncertainty into pandemic discourse around the world.

And finally, governmental communications about COVID-19 included inconsistent and contradictory narratives emerging from political leaders and institutions that should have been cooperating and coordinating.

As these three uncertainties played out last year, the public’s response to the coronavirus emergency fractured along ideological lines, with partisanship often replacing science as the justification for public health measures. Unfortunately, the internet-fueled undercutting of rational discourse and policy making is not specific to COVID-19. Efforts to deal with the existential threats of nuclear war and climate change have been similarly undermined.

Social media, search engines, always-on mobile computing technologies, and other technology applications have exploited human cognitive propensities to be misled and enraged and to react impulsively, exacerbating political and ideological differences. Established institutions that have traditionally provided a trusted center that supports societal stability—government agencies, especially those related to public health and climate change, journalism, the judiciary, education—are under attack precisely because they have provided stability.

At the very least, the widespread dysfunction in today’s information ecosystem is a threat multiplier that vastly complicates society’s ability to address major challenges. Pandemic responses in some countries, including the United States, have provided graphic demonstrations that such concerns are not merely theoretical. Disinformation has led leaders and citizens alike to reject scientific advice about limiting the spread of COVID-19, with tragic results.



One of many “mythbuster” videos and graphics that the WHO has produced in an effort to combat pandemic mis- and disinformation: Spraying and introducing bleach into your body WON'T protect you. (Source: World Health Organization)

Unchecked internet disinformation could have even more drastic consequences in a nuclear crisis, perhaps leading to a nuclear war that ends world civilization. Disinformation efforts across communications systems are at this moment undermining responses to climate change in many countries. The need for deep thinking and careful, effective action to counter the effects of internet-enabled disinformation has never been clearer.

Meanwhile, the COVID-19 pandemic continues to rage. SARS-CoV-2 took advantage of both physiological and societal vulnerabilities and continues to nimbly skirt poorly mustered defenses. Recent mutations have created variants of the virus that are more infectious and sicken children, who were previously thought to be less prone to infection.

Scientists around the world have mobilized to create COVID-19 treatments and vaccines, and their work is showing promise in reducing the severity of and eventually suppressing the pandemic. But public officials who have dismissed the value of science during the pandemic now face populations hesitant to take COVID-19 vaccines. Those same public figures also failed to iron out the manufacturing, distribution, and other logistical details needed for efficient immunization programs.

As this pandemic subsides, leaders around the world must come together to create the institutions and surveillance regimes that can identify disease outbreaks and quash them before they become pandemics, quickly develop vaccines and therapeutics for new diseases, and rapidly promulgate preventive measures for public health.

Rapidly advancing biological research and development have produced, and will continue to produce, disruptive technologies that could increase biological risk. In the risk-increasing category are biotechnology applications that could, for example, create super-soldiers or produce biological weapons. Many countries and corporations are investing in the biological sciences as they recognize the immense opportunities to establish and grow bioeconomies. These bio-investment programs raise the new possibility that nations may conduct biological weapons research and development under the guise of building effective responses for naturally occurring pandemics.

Bad actors have surely taken notice of the gaps in national responses to the COVID-19 pandemic. Most nations were unable to meet needs for personal protective equipment, to provide enough hospital beds to treat everyone who became seriously ill with the disease, or to manage international supply chains well enough to deliver medicines and equipment when and where they were most needed. International security requires speedy action to reduce those vulnerabilities. An improved global public health effort to prevent, detect, respond to, and recover from natural pandemics would, as a salutary side-effect, better prepare the world to respond to biological accidents and attacks.


This is your wake-up call

When the world finally emerges from the worst pandemic in a century, everyone will rightly celebrate. It might be tempting to mark the COVID-19 experience as a one-off, a dismal anomaly to be forgotten. We, too, wish the world could return to normal in short order.

But the pandemic is not a unique departure from a secure reality. It is a harbinger, an unmistakable signal that much worse will come if leaders and institutions do not enact wide-ranging reforms to forestall and minimize future pandemics, to restore the primacy of science-based policies, and to reduce the possibility of nuclear war and the impacts of climate change.

We set the Doomsday Clock at 100 seconds to midnight—the closest it has ever been—because the existential risks confronting humanity today call for quick and comprehensive action across the 21st century’s complex threat spectrum. Here are some practical steps that world leaders can and should initiate in 2021 to protect humanity from major global threats that have the potential to end civilization:
The US and Russian presidents should, upon extension of New START, launch follow-on talks for more ambitious and comprehensive limits of nuclear weapons and delivery systems.
Now that the United States has announced it will rejoin the Paris climate agreement, it should accelerate its commitment to decarbonization and put policies in place that make the attainment of the commitment feasible.
Now that the United States has rejoined the World Health Organization, it should work through the WHO and other international institutions to reduce biological risks of all kinds. Also, national leaders and international organizations can prepare for biological events before they occur by more carefully monitoring animal-human interactions and improving international disease surveillance and reporting efforts; increasing world capacity to produce and quickly distribute medical supplies; and expanding hospital capacity.
US President Joe Biden can show leadership by reducing US reliance on nuclear weapons via limits on their roles, missions, and platforms, and by decreasing budgets accordingly. The United States should declare its commitment to no-first-use of nuclear weapons and persuade allies and rivals to agree that no-first-use is a step toward security and stability.
President Biden should banish the fear that a single person would have the power to end civilization by eliminating his own and future US presidents’ sole authority to launch nuclear weapons. He should work to persuade other countries with nuclear weapons to put in place similar barriers.
Russia can rejoin the NATO-Russia Council and open serious discussions on risk reduction and on avoiding escalation dangers.
North Korea can agree to codify and allow verification of its moratorium on nuclear tests and long-range missile tests.
Iran and the United States can jointly return to full compliance with the Joint Comprehensive Plan of Action, and Iran can agree to new, broader talks about Middle East security and constraints on its missile and other military activities.
The United States and Russia can renew cooperation on fissile material and nuclear security to make sure that terrorists cannot acquire the means to build a nuclear weapon.
Banks and other sources of capital can implement policies that limit investment in fossil fuel projects, as indeed some already have done, and redirect it to climate-friendly investments.
China can reorient its Belt and Road Initiative, so it sets an example for other investors by pursuing sustainable development pathways rather than supporting fossil fuel-intensive development.
All nations can commit to stronger decarbonization goals under the Paris Agreement and implement policies directed toward the realization of these goals. Those policies should address not merely long-term goals but near-term emission reductions and investments in longer-term structural changes. Meanwhile, the world’s wealthier countries should enhance their commitments under the Paris Agreement to provide financial support and technology cooperation required by developing countries to undertake strong climate action.
Leaders in governments and the private sector can emphasize COVID-recovery investments that strongly favor climate mitigation and adaptation objectives across all economic sectors and address the full range of potential greenhouse gas emission reductions. This includes capital investments in urban development, agriculture, transport, heavy industry, buildings and appliances, and electric power.
The new US administration can fill leadership positions for science-based agencies on the basis of scientific expertise and credentials; prohibit interference with the production or dissemination of executive branch scientific reports; use the best possible science to inform policy considerations; allow government scientists to engage with the public about their work; and provide funding to restore and strengthen international scientific cooperation.
National leaders and international organizations can create more effective regimes for monitoring biological research and development efforts, so potential benefits can be maximized, and possible negative consequences minimized or eliminated.
Governments, major communications technology firms, academic experts, and responsible media organizations can cooperate to find practical and ethical ways to combat internet-enabled misinformation and disinformation.

Having now killed more than two million human beings, COVID-19 is an unmistakable global wake-up call. The message is simple and chilling: Next time could be far worse. Given the pandemic experience, no one can reasonably say he or she was not warned. It remains 100 seconds to midnight, the most dangerous situation that humanity has ever faced. It is time for all to take the actions needed to—quite literally—save the world.


About the Bulletin of the Atomic Scientists


At our core, the Bulletin of the Atomic Scientists is a media organization, publishing a free-access website and a bimonthly magazine. But we are much more. The Bulletin’s website, iconic Doomsday Clock, and regular events equip the public, policymakers, and scientists with the information needed to reduce manmade threats to our existence. The Bulletin focuses on three main areas: nuclear risk, climate change, and disruptive technologies. What connects these topics is a driving belief that because humans created them, we can control them. The Bulletin is an independent, nonprofit 501 (c) (3) organization. We gather the most informed and influential voices tracking man-made threats and bring their innovative thinking to a global audience. We apply intellectual rigor to the conversation and do not shrink from alarming truths.

The Bulletin has many audiences: the general public, which will ultimately benefit or suffer from scientific breakthroughs; policymakers, whose duty is to harness those breakthroughs for good; and the scientists themselves, who produce those technological advances and thus bear a special responsibility. Our community is international, with half of our website visitors coming from outside the United States. It is also young. Half are under the age of 35.


Learn more at thebulletin.org/about-us.

Boeing gets OK for satellite grid to provide internet from space

A NASA TV video frame grab shows the SpaceX Falcon 9 fourth Starlink constellation after entry burn, before it separates, after
A NASA TV video frame grab shows the SpaceX Falcon 9 fourth Starlink constellation 
after entry burn, before it separates, after it launched at Cape Canaveral, Florida on 
January 29, 2020. Boeing has just won approval for its satellite constellation to provide 
internet services from space.

Boeing on Wednesday gained US authorization for a project to launch satellites that will provide internet services from space.

The Federal Communications Commission (FCC) said in a statement it had approved a license for the aerospace giant "to construct, deploy, and operate a satellite constellation" that will "provide broadband and communications services for residential, commercial, institutional, governmental, and professional users in the United States and globally."

"Advanced satellite broadband services have an important role to play in connecting hard-to-serve communities," said Jessica Rosenworcel, the FCC chairwoman.

The FCC gave the  for 147 satellites, the vast majority of which will be in low orbit: 132 could be placed at an altitude of about 600 miles (1,000 km), and 15 would be much higher, between about 17,000 and 27,000 miles.

The service will first be available to clients in the United States and then around the world.

"Boeing sees a multi-orbit future for satellite technologies," the aerospace company said in a statement.

"As the demand for  grows, diversity will be required across orbital regimes and frequencies to satisfy unique customer demands, and we see V-band as helping to provide some of that diversity," Boeing added.

Other satellite constellation projects are already being rolled out by competing companies.

US billionaire Elon Musk, head of the space company SpaceX, has already put more than 1,500 satellites into orbit to create the Starlink network, while Amazon founder Jeff Bezos has a similar project called Kuiper.

© 2021 AFP

A framework to automatically identify wildlife in collaboration with humans

A framework to automatically identify wildlife in collaboration with humans
In real-world applications, AI models do not stop at one training stage. As data 
collection progresses over time, there is a continuous cycle of inference, 
annotation, and model updating. When there are novel and difficult samples, 
human annotation is inevitable. Credit: Miao et al.

Over the past few decades, computer scientists have developed numerous machine learning tools that can recognize specific objects or animals in images and videos. While some of these techniques have achieved remarkable results on simple animals or items (e.g., cats, dogs, houses), they are typically unable to recognize wildlife and less renowned plants or animals.

Researchers at University of California, Berkeley (UC Berkeley) have recently developed a new wildlife identification approach that performs far better than techniques developed in the past. The approach, presented in a paper published in Nature Machine Intelligence, was conceived by Zhongqi Miao, who initially started exploring the idea that artificial intelligence (AI) tools could classify wildlife images collected by movement-triggered . These are cameras that wildlife ecologists and researchers often set up to monitor species inhabiting specific geographic locations and estimate their numbers.

The  of AI for identifying species in wildlife images captured by camera traps could significantly simplify the work of ecologists and reduce their workload, preventing them from having to look through hundreds of thousands of images to generate maps of the distribution of species in specific locations. The framework developed by Miao and his colleagues is different from other methods proposed in the past, as it merges machine learning with an approach dubbed 'humans in the loop' to generalize better on real-world tasks.

"An important aspect of our 'humans in the loop innovation' is that it addresses the 'long-tailed distribution problem," Wayne M. Getz, one of the researchers who carried out the study, told TechXplore. "More specifically, in a set of hundreds of thousands of images generated using camera traps deployed in an area over a season, images of common species may appear hundreds or even thousands of times, while those of rare species may appear just a few times. This produces a long-tailed distribution of the frequency of images of different species."

If all species were captured by camera traps with equal frequency, their distribution would be what is known as 'rectangular." On the other hand, if these frequencies are highly imbalanced, the most common frequencies (plotted first down the y-axis) would be far larger than least common frequencies (plotted at the bottom of the graph), resulting in a long-tailed distribution

"If standard AI image recognition software were applied to long-tailed distributional data, then the method would fail miserably when it comes to identifying rare species," Getz explained. "The primary purpose of our study was to find a way to improve the identification of rare species by incorporating humans into the process in an iterative manner."

When trying to apply conventional AI tools in real-world settings, computer scientists can encounter several challenges. As mentioned by Getz, the first is that data collected in the real world often follows a long-tail distribution and current state-of-the-art AI models do not perform as well on this data, compared to data with a rectangular or normal distribution.

"In other words, when applied to data with a long-tailed distribution, large or more frequent categories always lead to much better performance than smaller and rare categories," Miao, lead author of the paper, told TechXplore. "Furthermore, instances of rare categories (especially images of rare animals) are not easy to collect, making it even harder to get around this long-tail distribution issue through data collection."

Another challenge of applying AI in real-world settings is that the problems they are meant to solve are typically open-ended. For instance, wildlife monitoring projects can continue indefinitely and span across long periods of time, during which new camera traps will be set up and a variety of new data will be collected.

In addition, new animal species might suddenly appear in the sites monitored by the cameras due to several possible factors, including unexpected invasions, animal reintroduction projects or recolonizations. All of these changes will be reflected in the data, ultimately impairing the performance of pre-trained machine learning techniques.

"So far, the human contribution to the training of AI has been inevitable," Miao said. "As real-world applications are open-ended, ensuring that AI models learn and adapt to new content requires additional human annotations, especially when we want the models to identify new animal species. Thus, we think there is a loop of AI recognition system of new data collection, human annotation on new data and model update to the novel categories."

In their previous research, the researchers tried to address the factors impairing the performance of AI in real-world settings in several different ways. While the approaches they devised were in some ways promising, their performance was not as good as they had hoped, achieving a classification accuracy below 70 percent when tested on standardized long-tailed datasets.

"It's hard for people to trust an AI model that could only produce ~70 percent accuracy," Miao said. "Overall, we think a deployable AI model should: achieve a balanced performance across imbalanced distribution (long-tailed recognition), be able to adapt to different environments (multi-domain adaptation), be able to recognize novel samples (out-of-distribution detection), and be able to learn from novel samples as fast as possible (few-shot learning, life-long learning, etc.). However, each one these characteristics have proved difficult to realize, and none of them have been fully solved yet, let alone combining them together and coming up with a perfect AI solution."

Instead of using renowned and existing AI tools or trying to develop an 'ideal' method, therefore, Miao and his colleagues decided to create a highly performing tool that relies on a certain amount of input from humans. As so far human annotations on data have proved to be particularly valuable for enhancing the performance of deep learning-based models, they focused their efforts on maximizing their efficiency.

"The goal of our project was to minimize the need for human intervention as much as possible, by applying human annotation solely on difficult images or novel species, while maximizing the recognition performance/accuracy of each model update procedure (i.e., update efficiency)," Miao said.

By combining machine learning techniques with human efforts in an efficient way, the researchers hoped to achieve a system that was better at recognizing animals in real-world wildlife images, overcoming some of the issues they encountered in their past studies. Remarkably, they found that their method could achieve 90 percent accuracy on wildlife image classification tasks, using 1/5 of the annotations that standard AI approaches would require to achieve this accuracy.

"Putting AI techniques into practice has always been significantly challenging, no matter how promising theoretical results are in previous studies on standard datasets," Miao said. "We thus tried to propose an AI recognition framework that can be deployed in the field even when the AI models are not perfect. And our solution is to introduce efficient human efforts back into the recognition system. And in this project, we use wildlife recognition as a practical use case of our framework."

Instead of evaluating AI models using a single dataset, the framework devised by Miao and his colleagues focuses on how efficiently a previously trained model can analyze newly collected datasets containing images of previously unobserved species. Their approach incorporates an active learning technique, which uses a prediction confidence metric to select low-confidence predictions, so that they can be annotated further by humans. When a model identifies animals with high levels of confidence, on the other hand, their framework stores these predictions as pseudo labels.

"Models are then updated according to both human annotations and pseudo labels," Miao explained. "The model is evaluated based on: the overall validation accuracy of each category after the update (i.e., update performance); percentage of high-confidence predictions on validation (i.e., saved human effort for annotation); accuracy of high-confidence predictions; and percentage of novel categories that are detected as low-confidence predictions (i.e., sensitivity to novelty)."

The overall aim of the optimization algorithm used by Miao and his colleagues is to minimize human efforts (i.e., to maximize a model's high-confidence percentage), while maximizing performance and accuracy. Technically speaking, the researchers' framework is a combination of active learning and semi-supervised learning with humans in the loop. All of the codes and data used by Miao and his colleagues are publicly available and can be accessed online.

"We proposed a deployable human-machine recognition framework that is also applicable when the models are not perfectly performing by themselves," Miao said. "With the iterative human-machine updating procedure, the framework can keep updated be deployed when new data are continuously collected. Furthermore, each technical component in this framework can be replaced with more advanced methods in the future to achieve better results."

The experimental setting outlined by Miao and his colleagues is arguably more realistic than those considered in previous works. In fact, instead of focusing on a single cycle of model training, validation and testing, it focuses on numerous cycles or stages, which allows models to better adapt to changes in the data.

"Another unique aspect of our work is that we proposed a synergistic relationship between humans and machines," Miao said." Machines help relieve the burden of humans (e.g., ~80 percent annotation requirements), and humans help annotate novel and challenging samples, which are then used to update the machines, such that the machines are more powerful and more generalized in the future. This is a continuous and long-term relationship."

In the future, the framework introduced by this team of researchers could allow ecologists to monitor animal species in different places more efficiently, reducing the time they spend examining images collected by trap cameras. In addition, their framework could be adapted to tackle other real-world problems that involve the analysis of data with a long-tailed distribution or that continuously changes over time.

"Miao is now working on the problem of trying to identify species from satellite or aerial images which present two challenges compared with camera trap images: the resolution is much lower because cameras are much more distant from the subjects that are capturing and the individual being imaged may be one of many in the overall frame; images generally show only a 1-d projection (i.e., from the top) rather than the 2-d projections (front/back and leftside/rightside) of camera trap data," Getz said.

Miao, Getz ad their colleagues now also plan to deploy and test the framework they created in real-world settings, such as camera trap wildlife monitoring projects in Africa organized by some of their collaborators. Meanwhile, Miao is working on other deep learning tools for the analysis of aerial images and audio recordings, as these could be particularly useful for identifying birds or marine animals. His overall goal is to make deep learning more accessible for ecologists and researchers analyzing wildlife images.

"On a broader scale, we think that the synergistic relationship between humans and machines is an exciting topic and that the goal of AI research should be to develop tools that augment people's abilities (or intelligence), rather than to eliminate the existence of humans (e.g., looking for perfect machines that can handle everything without the need for humans)," Miao added. "It is more like a loop where machines make humans better, and humans make machines more powerful in return, just like in the iterative framework we proposed in the paper. We call this Artificial Augmented Intelligence (A2I or A-square I), where ultimately, people's intelligence is augmented with artificial intelligence and vice versa. In the future, we want to keep exploring the possibilities of A2I."Researchers successfully train computers to identify animals in photos

More information: Zhongqi Miao et al, Iterative human and automated identification of wildlife images, Nature Machine Intelligence (2021). DOI: 10.1038/s42256-021-00393-0

Ziwei Liu et al, Large-scale long-tailed recognition in an open world. arXiv:1904.05160v2 [cs.CV], arxiv.org/abs/1904.05160

Ziwei Liu et al, Open compound domain adaptation. arXiv:1909.03403v2 [cs.CV], arxiv.org/abs/1909.03403

Journal information: Nature Machine Intelligence 

Using ocean plastic waste to power ocean cleanup ships

Using ocean plastic waste to power ocean cleanup ships
Proposals for ocean plastic cleanup currently require traveling back to port to 
unload the plastics and refuel the vessel. Credit: Worcester Polytechnic Institute.

A team of researchers from Worcester Polytechnic Institute, Woods Hole Oceanographic Institution and Harvard University believes that the plastic amassing in floating islands in the oceans could be used to power the ships that are sent to clean them up. In their paper published in Proceedings of the National Academy of Sciences, the group describes how ocean plastics could be converted to ship fuel.

Prior research has shown that millions of tons of plastics enter the  each year—some of it is ground into fragments and disperses, and some of it winds up in colossal garbage patches floating in remote parts of the ocean. Because of the danger that such plastics present to ocean life, some environmentalists have begun cleanup operations. Such operations typically involve sending a ship to a garbage patch, collecting as much as the ship will hold and then bringing it back to port for processing. In this new effort, the researchers suggest it would be far more efficient and greener to turn the  into fuel for both a processing machine and for uninterrupted operation of the ships.

The researchers note that the plastic in a  could be converted to a type of oil via hydrothermal liquefaction (HTL). In this process, the plastic is heated to 300–550 degrees Celsius at pressures 250 to 300 times that of sea-level conditions. The researchers have calculated that a ship carrying an HTL converter would be capable of producing enough oil to run the HTL converter and the ship's engine. Under their scenario, plastic collection booms would be permanently stationed at multiple sites around a large  patch, able to load the plastic it collects onto ships.

The researchers acknowledge that burning the oil produced would release carbon into the atmosphere, but note that the amount emitted would still be less than that emitted by a ship burning conventional oil making trips back and forth to ports. They also note that HTL does produce a small amount of solid waste, which would have to be taken back to port, likely every few months—excess fuel produced by the HTL could be used for these trips.

'The Ocean Cleanup' ship sweeps first Pacific plastic

More information: Elizabeth R. Belden et al, Thermodynamic feasibility of shipboard conversion of marine plastics to blue diesel for self-powered ocean cleanup, Proceedings of the National Academy of Sciences (2021). DOI: 10.1073/pnas.2107250118

Journal information: Proceedings of the National Academy of Sciences 

© 2021 Science X Network

Hybrid cars' green credentials under scrutiny

Sales of hybrid cars, which use both a conventional combustion engine and a small electric motor, could soon overtake those of p
Sales of hybrid cars, which use both a conventional combustion engine and a 
small electric motor, could soon overtake those of petrol vehicles in the EU.

Hybrid cars are increasingly popular in the European Union as eco-conscious drivers turn away from their more polluting petrol and diesel counterparts, but environmentalists warn they're not as green as they seem.

Sales of the cars, which use both a conventional combustion engine and a small electric motor, allowing owners to drive a few kilometres without emitting CO2, could soon overtake those of petrol vehicles in the EU.

In the third quarter of this year, 20.7 percent of cars sold in the bloc were new hybrid versions whose batteries are recharged by collecting wasted energy from elsewhere, like braking, and 9.1 percent were hybrid plug-ins that can be charged from an electric outlet.

Close to 40 percent were petrol-powered, 17.6 percent diesel and just 9.8 percent were fully electric.

Cheaper than fully , they also provide some reassurance for those worried about their battery running out of power at a time when charging stations are still not widespread.

Auto giants like Toyota, Stellantis, Renault and Hyundai-Kia are banking on hybrids, not least because they allow them to comply with EU norms on CO2 emissions at a lesser cost than fully electric cars.

'Barely cleaner'

But are they truly less polluting, or more of a transition solution as the world edges towards ditching petrol and diesel altogether?

Greenpeace and the pressure group Transport & Environment believe that hybrids actually slow down this transition.

They want to accelerate the shift to fully electric and to other forms of transport, pointing out that hybrids aren't that green.

"Conventional 'full' hybrids in particular, which run for the majority of the time on fossil fuel energy, are barely any cleaner than traditional petrol and diesel engines," Greenpeace said last year.

Marie Cheron of France's Nicolas Hulot Foundation, an environmental group, concurred.

"For example, some hybrids have been bought for fleets (of cars), they do not have a system that allows them to recharge, people don't charge them, and so they don't drive electric."

But Philippe Degeilh, an engineer at IFP Energies Nouvelles (Ifpen), an energy, transport and environment research group, said people just need to be educated in how to use hybrids correctly.

According to an Ifpen study published at the end of 2020, hybrids emit an average of 12 percent less CO2 than a similar petrol-powered car.

That rises to 33 percent in town, while it drops to almost zero on highways.

Plug-ins that are driven smoothly—draining batteries less—and often recharged are "capable of nearing zero emissions," according to Ifpen.

"A household that has just one car can have a better environmental record with a hybrid rather than with an electric car equipped with a large battery. It's designed to do 50 kilometres a day and sometimes to go on holiday," said Degeilh.

To stay or not?

Meanwhile, fully electric cars aren't necessarily all that green either.

Their batteries, which are getting bigger and bigger, require a lot of energy in their production.

Where the electricity comes from is also important to determine their environmental credentials.

The debate around hybrids is also a political one.

As the EU plans to ban the sale of petrol and diesel engines from 2035, some of the auto industry wants to ensure a role for hybrids.

"We think the  is here to stay," Jim Crosbie, head of Toyota Motor Manufacturing France, told AFP.

Hybrids—excluding plug-ins—represent 70 percent of the Japanese group's sales in Western Europe.

"If we're talking about a model life cycle of seven to nine years, it will remain an important asset for us in the years to come," he said.Sales of electric cars charge ahead in Europe

© 2021 AFP

Augmented reality: an early taste of the metaverse?

Under Peggy Johnson, Magic Leap has pivoted to developing augmented reality goggles for professionals, including surgeons
Under Peggy Johnson, Magic Leap has pivoted to developing augmented reality
 goggles for professionals, including surgeons.

When Facebook unveiled a mock-up last week of the "metaverse"—supposedly the internet of the future—it showed people transported to a psychedelic world of flying fish and friendly robots.

But while even Facebook CEO Mark Zuckerberg acknowledges these kinds of experiences could be many years away, some enthusiasts argue that a more modest version of the metaverse is already here.

"We're in the early stages of the metaverse, in some ways," Peggy Johnson, CEO of Magic Leap, told AFP at the Web Summit in Lisbon on Tuesday.

Magic Leap makes augmented reality (AR) headsets, which have already been used by surgeons preparing to separate a pair of conjoined twins, and by factory supervisors carrying out site inspections.

In both cases, information popped up before the users' eyes about what they were seeing.

It might not feel quite as immersive—or as kooky—as the virtual reality (VR) experiences that Zuckerberg wants to eventually bring to people's homes. But it nonetheless blurs the divide between the physical world and the digital one, a key idea behind the metaverse.

"With VR, you put on a device, and then you're in another world," Johnson said. "With AR, you put on a device, you're still in your world, but we're augmenting it with digital content."

So far, many people's experiences of AR have been limited to playing Pokemon Go or experimenting with image filters that transplant a comical pair of ears onto someone's face.

But it is in healthcare that the true potential of AR is starting to be realised, Johnson said.

Magic Leap's first augmented reality headset, released in 2018, failed to take off among the general public
Magic Leap's first augmented reality headset, released in 2018, failed to take off
 among the general public.

"You can call in experts who can look at the same thing as you are, from another part of the world," she said. "During surgery, you can lay down digital lines where perhaps the incision is going to occur."

Founded in 2010, Magic Leap's initial mission to bring AR to the masses generated huge hype and nearly $2.3 billion in venture funding.

Early promo material imagined it being used to bring a killer whale into a gymnasium full of schoolchildren.

But when Magic Leap's first headset was finally revealed in 2018, there was widespread disappointment; the product was too bulky and expensive to catch on among the general public.

The company was forced to lay off around half its staff last year.

Restaurant reviews and forgotten names

Johnson, a former Microsoft executive, took over as CEO in August 2020 and pivoted towards developing the goggles for use by professionals.

The Florida-based company last month announced that it has raised another $500 million in funding, with a new headset, the Magic Leap 2, set to be released in 2022.

The updated version is more lightweight, but it is still set to be used mostly by people accustomed to wearing goggles at work—like surgeons performing delicate work, or defence industry specialists.

If the AR revolution arrives, the market may be a crowded one with companies like Snapchat's developer, Snap, trialling spectacl
If the AR revolution arrives, the market may be a crowded one with companies
 like Snapchat's developer, Snap, trialling spectacles.

Google Glass, a pair of "smart glasses" that failed to take off when they launched in 2014, has similarly re-emerged as a product aimed at professional users.

Johnson predicted it might still be "a few more years" before Magic Leap or one of its competitors creates an AR headset that could feasibly be worn by consumers everywhere.

But that's the moment when Johnson predicts that AR could really transform our everyday lives.

It might, she suggested, allow us to see reviews for restaurants pinging before our eyes as we walk down a street perusing the options.

Forgotten someone's name? No problem. As they walk towards you, it could appear above their head.

"Right now we're all looking down at our mobile phones," Johnson said. Augmented reality, she hopes, could help us to soak up the world around us—a world with extra information layered over the top of it.

If that revolution arrives, the market may be a crowded one. Facebook is working on its own AR headset, while Apple is rumoured to be following suit. Snapchat's developer, Snap, is meanwhile trialling a new pair of its "Spectacles" on AR artists.

What does Johnson think the metaverse will look like in 15 years?

"I think you'll go back home to pick up your glasses because you left them at home," she predicted. "The same way you do with your mobile phone today."

Facebook assembles team to build 'metaverse'

© 2021 AFP

Study finds public support for nuclear energy in Southeast Asia generally low

nuclear plant
Credit: CC0 Public Domain

Nuclear energy may be the world's second-largest low carbon energy source for generating electricity after hydroelectric power, but reception to its adoption remains lukewarm in Southeast Asia, an NTU Singapore study has found.

Conducted by NTU's Wee Kim Wee School of Communication and Information, the study surveyed 1,000 people each in Singapore, Malaysia, Indonesia, Vietnam, and Thailand through door-to-door questionnaires and found that more than half of the respondents in every country were against the idea of  energy development.

Based on its surveys, the study found that about one in five (22 percent) of those surveyed in Singapore were in favor of nuclear energy development. The level of  in the other four countries surveyed ranged from 3 percent to 39 percent.

The NTU scientists also found that the respondents tended to use "cognitive shortcuts" such as risk and benefit perception (an individual's belief in the threat or benefits of nuclear energy), religious beliefs, and trust in various entities such as university scientists, business leaders, and the government to aid their decision on their level of support for nuclear  development.

With the five countries surveyed in this NTU study geographically close to each other, having a nuclear power plant in any of the five Southeast Asian countries will impact the others, said Prof. Shirley Ho, who led the study.

Prof. Ho, who is NTU's Research Director for Arts, Humanities, Education, and Social Sciences, added that the findings are a key point for consideration for policymakers in these countries, given that data suggests the public is collectively unsupportive of having a nuclear power plant in their own country.Merkel: No way back on German plan to end nuclear power use

More information: The study is available as a PDF at www.ntu.edu.sg/docs/default-so … df?sfvrsn=6d2991a3_1

Provided by Nanyang Technological University 

'Trojan Source' bug a novel way to attack program encodings

hacked data
Credit: CC0 Public Domain

A pair of security experts at TrojanSource have found a novel way to attack computer source code—one that fools a compiler (and human reviewer) into thinking code is safe. Nicholas Boucher and Ross Anderson, both with the University of Cambridge, have posted a paper on the TrojanSource web page detailing the vulnerability and ways that it might be fixed.

As Boucher and Anderson describe it, the vulnerability involves  being committed by nefarious types using Unicode control characters to reorder characters in source  that appears to programmers to be legitimate. More specifically, the vulnerability involves the use of a 'Bidi' algorithm, in Unicode (an international encoding standard that can be used in ) where characters can be placed both left to right and right to left—because some languages, such as Hebrew and Arabic are written and read right to left.

The vulnerability exists because the algorithms that process such code do not take into consideration that some of the characters that are being read left to right, can have a different meaning or purpose if they are read right to left. Because virtually all of the most popular programming languages in use today—C, C+, Java, Python, Go, Rust and JavaScript—allow Unicode, that means that virtually all programs are potentially at risk.

As an example, Boucher and Anderson show that a line of code such as:

/* begin admins only */ if (isAdmin) {

Could be changed to:

/* if (isAdmin) { begin admins only */

The first line is a harmless comment inserted by a programmer, the second is code that could be used to conduct a desired outcome by a hacker. The researchers suggest the vulnerability represents a serious threat to software supply chains—if such vulnerabilities were exploited, they could impact downstream software by allowing them to inherit the same vulnerability.

Because the  exists for such a wide variety of programming languages, its disclosure was first coordinated with officials charged with maintaining the rules for such languages giving them time to add changes to compilers and interpreters to account for and mitigate such a threat.

Vulnerability found in Kindle e-reader

More information: Report: www.trojansource.codes/trojan-source.pdf

TrojanSource: www.trojansource.codes/

© 2021 Science X Network