AI has potential benefits and potential risks.[192] AI may be able to advance science and find solutions for serious problems: Demis Hassabis of DeepMind hopes to “solve intelligence, and then use that to solve everything else”. However, as the use of AI has become widespread, several unintended consequences and risks have been identified.[195] In-production systems can sometimes not factor ethics and bias into their AI training processes, especially when the AI algorithms are inherently unexplainable in deep learning.
Risks and harm
Privacy and copyright
Machine learning algorithms require large amounts of data. The techniques used to acquire this data have raised concerns about privacy, surveillance and copyright.
AI-powered devices and services, such as virtual assistants and IoT products, continuously collect personal information, raising concerns about intrusive data gathering and unauthorised access by third parties. The loss of privacy is further exacerbated by AI’s ability to process and combine vast amounts of data, potentially leading to a surveillance society where individual activities are constantly monitored and analysed without adequate safeguards or transparency.
Sensitive user data collected may include online activity records, geolocation data, video, or audio. For example, in order to build speech recognition algorithms, Amazon has recorded millions of private conversations and allowed temporary workers to listen to and transcribe some of them. Opinions about this widespread surveillance range from those who see it as a necessary evil to those for whom it is clearly unethical and a violation of the right to privacy.
AI developers argue that this is the only way to deliver valuable applications and have developed several techniques that attempt to preserve privacy while still obtaining the data, such as data aggregation, de-identification and differential privacy. Since 2016, some privacy experts, such as Cynthia Dwork, have begun to view privacy in terms of fairness. Brian Christian wrote that experts have pivoted “from the question of ‘what they know’ to the question of ‘what they’re doing with it’.”
Generative AI is often trained on unlicensed copyrighted works, including in domains such as images or computer code; the output is then used under the rationale of “fair use“. Experts disagree about how well and under what circumstances this rationale will hold up in courts of law; relevant factors may include “the purpose and character of the use of the copyrighted work” and “the effect upon the potential market for the copyrighted work”.[203] Website owners can indicate that they do not want their content scraped via a “robots.txt” file.[204] However, some companies will scrape content regardless[205][206] because the robots.txt file has no real authority. In 2023, leading authors (including John Grisham and Jonathan Franzen) sued AI companies for using their work to train generative AI. Another discussed approach is to envision a separate sui generis system of protection for creations generated by AI to ensure fair attribution and compensation for human authors.[209]
Dominance by tech giants
The commercial AI scene is dominated by Big Tech companies such as Alphabet Inc., Amazon, Apple Inc., Meta Platforms, and Microsoft.[210][211][212] Some of these companies already own the vast majority of existing cloud infrastructure and computing power from data centres, allowing them to entrench further in the marketplace.[213][214] For most of 2025-2026, AI-related companies have also dominated stock market returns. According to a July 2024 report from S&P Global, investors had started questioning whether the enormous spending on AI infrastructure and data centers would generate sufficient profits, leading to increased volatility in technology and semiconductor stocks and to concerns about a potential AI bubble.[215] In a 2025 Economic Policy article, economists Anton Korinek and Jai Vipra described some AI related market sectors as highly concentrated.[216]
Power needs and environmental impacts
Technology companies have built electricity and artificial intelligence infrastructure to facilitate the AI boom of the 2020s. A 2025 report from the consulting firm McKinsey & Company estimated that by 2030, $2.7 trillion would be invested into AI infrastructure and data centres in the US, surpassing World War II’s Manhattan Project every month.[218]
In January 2024, the International Energy Agency (IEA) released Electricity 2024, Analysis and Forecast to 2026.[219] This is the first IEA report to make projections for data centres and power consumption by AI and cryptocurrency. The report states that power demand for these uses might double by 2026, with the additional power consumption equaling that of Japan.[220]
Power consumption by AI is responsible for an increase in fossil fuel use, and has delayed closings of obsolete, carbon-emitting coal energy facilities. A ChatGPT search involves 10 times as much electrical energy as a Google search.[221]
A 2024 Goldman Sachs research paper, AI Data Centers and the Coming US Power Demand Surge, found “US power demand (is) likely to experience growth not seen in a generation….” and forecasts that, by 2030, US data centres will consume 8% of US power, as opposed to 3% in 2022, presaging growth for the electrical power generation industry by a variety of means.[222] Data centres’ need for more and more electrical power is such that they might max out the electrical grid. The Big Tech companies counter that AI can be used to maximise the utilisation of the grid by all.[223]
In 2024, The Wall Street Journal reported that big AI companies have begun negotiations with the US nuclear power providers to provide electricity to the data centres. In March 2024, Amazon purchased a Pennsylvania nuclear-powered data centre for US$650 million.[224]
In September 2024, Microsoft announced an agreement with Constellation Energy to re-open the Three Mile Island nuclear power plant to provide Microsoft with 100% of all electric power produced by the plant for 20 years. Reopening the plant, which suffered a partial nuclear meltdown of its Unit 2 reactor in 1979, will require Constellation to get through strict regulatory processes which will include extensive safety scrutiny from the US Nuclear Regulatory Commission. If approved (this will be the first ever US re-commissioning of a nuclear plant), over 835 megawatts of power – enough for 800,000 homes – will be produced. The cost for re-opening and upgrading is estimated at US$1.6 billion and is dependent on tax breaks for nuclear power contained in the 2022 US Inflation Reduction Act.[225] As of 2024, the US government and the state of Michigan have been investing almost US$2 billion to reopen the Palisades nuclear reactor on Lake Michigan. Closed since 2022, the plant was scheduled to be reopened in October 2025.[226]
After the last approval in September 2023, Taiwan suspended the approval of data centres north of Taoyuan with a capacity of more than 5 MW in 2024, due to power supply shortages.[227] Taiwan aims to phase out nuclear power by 2025.[227]
Singapore imposed a ban on the opening of data centres in 2019 due to electric power, but lifted it in 2022.[227]
Although most nuclear plants in Japan have been shut down after the 2011 Fukushima nuclear accident, according to an October 2024 Bloomberg article in Japanese, cloud gaming services company Ubitus, in which Nvidia has a stake, is looking for land in Japan near a nuclear power plant for a new data centre for generative AI.[228]
On 1 November 2024, the Federal Energy Regulatory Commission (FERC) rejected an application submitted by Talen Energy for approval to supply some electricity from the nuclear power station Susquehanna to Amazon’s data centre.[229]
According to the Commission Chairman Willie L. Phillips, it is a burden on the electricity grid as well as a significant cost-shifting concern to households and other business sectors.[229]
In 2025, a report prepared by the IEA estimated the greenhouse gas emissions from the energy consumption of AI at 180 million tons. By 2035, these emissions could rise to 300–500 million tonnes depending on what measures will be taken. This is below 1.5% of the energy sector emissions. The emissions reduction potential of AI was estimated at 5% of the energy sector emissions, but rebound effects (for example if people switch from public transport to autonomous cars) can reduce it.[230]
In June 2026, the United Nations University Institute for Water, Environment and Health released a report about cumulative environmental impacts of AI, not only carbon. The report found that AI has significant negative impacts on land, water, and climate, as it uses large amounts of electricity, requires mining of minerals, and generates large amounts of electronic waste. In 2025, data centres consumed more electricity than Saudi Arabia, and ChatGPT alone consumed an amount of water equivalent to the needs of 500,000 people in Sub-Saharan Africa. The negative impacts are not distributed equally across the planet but are concentrated in some places. Data centres are often built in places already suffering from water deficit. For example, AI deployment has already led to an increase in water scarcity in Mexico and Uruguay. It created problems with energy supply in Ireland, which forbade construction of new data centres near Dublin until 2028. Measures can involve tradeoffs, for example, transitioning from coal to bioenergy reduces carbon footprint but increases the impact on land and water. Energy efficiency might not reduce environmental impact, because as AI becomes more efficient, it also becomes cheaper and more widely used.[11]
Misinformation
YouTube, Facebook and others use recommender systems to guide users to more content. These AI programs were given the goal of maximising user engagement (that is, the only goal was to keep people watching). The AI learned that users tended to choose misinformation, conspiracy theories, and extreme partisan content, and, to keep them watching, the AI recommended more of it. Users also tended to watch more content on the same subject, so the AI led people into filter bubbles where they received multiple versions of the same misinformation. This convinced many users that the misinformation was true, and ultimately undermined trust in institutions, the media and the government.[232] The AI program had correctly learned to maximise its goal, but the result was harmful to society. After the U.S. election in 2016, major technology companies took some steps to mitigate the problem.[233]
In the early 2020s, generative AI began to create images, audio, and texts that are virtually indistinguishable from real photographs, recordings, or human writing,[234] while realistic AI-generated videos became feasible in the mid-2020s.[235][236][237] It is possible for bad actors to use this technology to create massive amounts of misinformation and computational propaganda through techniques such as deepfakes. AI pioneer and Nobel Prize-winning computer scientist Geoffrey Hinton expressed concern about AI enabling “authoritarian leaders to manipulate their electorates” on a large scale, among other risks. The ability to influence electorates has been proved in at least one study. This same study shows more inaccurate statements from the models when they advocate for candidates of the political right.[241]
AI researchers at Microsoft, OpenAI, universities and other organisations have suggested using “personhood credentials” as a way to overcome online deception enabled by AI models.[242]
Algorithmic bias and fairness
Machine learning applications can be biased[k] if they learn from biased data. The developers may not be aware that the bias exists. Discriminatory behaviour by some LLMs can be observed in their output.[246] Bias can be introduced by the way training data is selected and by the way a model is deployed. If a biased algorithm is used to make decisions that can seriously harm people (as it can in medicine, finance, recruitment, housing or policing) then the algorithm may cause discrimination.[248] The field of fairness studies how to prevent harms from algorithmic biases.
On 28 June 2015, Google Photos‘s new image labelling feature mistakenly identified Jacky Alcine and a friend as “gorillas” because they were black. The system was trained on a dataset that contained very few images of black people, a problem called “sample size disparity”. Google “fixed” this problem by preventing the system from labelling anything as a “gorilla”. Eight years later, in 2023, Google Photos still could not identify a gorilla, and neither could similar products from Apple, Facebook, Microsoft and Amazon.
COMPAS is a commercial program widely used by U.S. courts to assess the likelihood of a defendant becoming a recidivist. In 2016, Julia Angwin at ProPublica discovered that COMPAS exhibited racial bias, despite the fact that the program was not told the races of the defendants. Although the error rate for both whites and blacks was calibrated equal at exactly 61%, the errors for each race were different—the system consistently overestimated the chance that a black person would re-offend and would underestimate the chance that a white person would not re-offend. In 2017, several researchers[l] showed that it was mathematically impossible for COMPAS to accommodate all possible measures of fairness when the base rates of re-offence were different for whites and blacks in the data.[254]
A program can make biased decisions even if the data does not explicitly mention a problematic feature (such as “race” or “gender”). The feature will correlate with other features (like “address”, “shopping history” or “first name”), and the program will make the same decisions based on these features as it would on “race” or “gender”.[255] Moritz Hardt said “the most robust fact in this research area is that fairness through blindness doesn’t work.”[256]
Criticism of COMPAS highlighted that machine learning models are designed to make “predictions” that are only valid if we assume that the future will resemble the past. If they are trained on data that includes the results of racist decisions in the past, machine learning models must predict that racist decisions will be made in the future. If an application then uses these predictions as recommendations, some of these “recommendations” will likely be racist.[257] Thus, machine learning is not well suited to help make decisions in areas where there is hope that the future will be better than the past. It is descriptive rather than prescriptive.[m]
Bias and unfairness may go undetected because the developers are overwhelmingly white and male: among AI engineers, about 4% are black and 20% are women.
There are various conflicting definitions and mathematical models of fairness. These notions depend on ethical assumptions, and are influenced by beliefs about society. One broad category is distributive fairness, which focuses on the outcomes, often identifying groups and seeking to compensate for statistical disparities. Representational fairness tries to ensure that AI systems do not reinforce negative stereotypes or render certain groups invisible. Procedural fairness focuses on the decision process rather than the outcome. The most relevant notions of fairness may depend on the context, notably the type of AI application and the stakeholders. The subjectivity in the notions of bias and fairness makes it difficult for companies to operationalise them. Having access to sensitive attributes such as race or gender is also considered by many AI ethicists to be necessary in order to compensate for biases, but it may conflict with anti-discrimination laws.[243]
At the 2022 ACM Conference on Fairness, Accountability, and Transparency a paper reported that a CLIP‑based (Contrastive Language-Image Pre-training) robotic system reproduced harmful gender‑ and race‑linked stereotypes in a simulated manipulation task. The authors recommended robot‑learning methods which physically manifest such harms be “paused, reworked, or even wound down when appropriate, until outcomes can be proven safe, effective, and just.”[259][260][261]
Lack of transparency
Many AI systems, particularly deep neural networks, are so complex that their designers cannot explain how they reach their decisions. An AI system which is not understandable is known as a “black box“.[263]
It is impossible to be certain that a program is operating correctly if no one knows how exactly it works. There have been many cases where a machine learning program passed rigorous tests, but nevertheless learned something different from what the programmers intended. For example, a system that could identify skin diseases better than medical professionals was found to actually have a strong tendency to classify images with a ruler as “cancerous”, because pictures of malignancies typically include a ruler to show the scale. Another machine learning system designed to help effectively allocate medical resources was found to classify patients with asthma as being at “low risk” of dying from pneumonia. Having asthma is actually a severe risk factor, but since the patients having asthma would usually get much more medical care, they were relatively unlikely to die according to the training data. The correlation between asthma and low risk of dying from pneumonia was real, but misleading.
People who have been harmed by an algorithm’s decision have a right to an explanation.[266] Doctors, for example, are expected to clearly and completely explain to their colleagues the reasoning behind any decision they make. Early drafts of the European Union’s General Data Protection Regulation in 2016 included an explicit statement that this right exists.[n] Industry experts noted that this is an unsolved problem with no solution in sight. Regulators argued that nevertheless the harm is real: if the problem has no solution, the tools should not be used.
DARPA established the XAI (“Explainable Artificial Intelligence”) program in 2014 to try to solve these problems.
Several approaches aim to address the transparency problem. SHAP enables to visualise the contribution of each feature to the output. LIME can locally approximate a model’s outputs with a simpler, interpretable model. Multitask learning provides a large number of outputs in addition to the target classification. These other outputs can help developers deduce what the network has learned. Deconvolution, DeepDream and other generative methods can allow developers to see what different layers of a deep network for computer vision have learned, and produce output that can suggest what the network is learning. For generative pre-trained transformers, Anthropic developed a technique based on dictionary learning that associates patterns of neuron activations with human-understandable concepts.[273]
Bad actors
Artificial intelligence provides a number of tools that are useful to bad actors, such as authoritarian governments, terrorists, criminals or rogue states. There are many ways in which AI is expected to help bad actors, some of which cannot be foreseen. For example, machine-learning AI is able to design tens of thousands of toxic molecules in a matter of hours.
Authoritarian governance
AI tools make it easier for governments to efficiently control their citizens and can increase digital authoritarianism and political repression.[275][276] Face and voice recognition allow widespread surveillance. Machine learning, operating this data, can classify potential enemies of the state and prevent them from hiding. Recommendation systems can precisely target propaganda and misinformation for maximum effect. Deepfakes and generative AI aid in producing misinformation. Advanced AI can make authoritarian centralised decision-making more competitive than liberal and decentralised systems such as markets. It lowers the cost and difficulty of digital warfare and advanced spyware. AI facial recognition systems are already being used for mass surveillance in China.[278][279]
Technological unemployment
Economists have frequently highlighted the risks of redundancies from AI, and speculated about unemployment if there is no adequate social policy for full employment.
The Economist said in 2015 that “the worry that AI could do to white-collar jobs what steam power did to blue-collar ones during the Industrial Revolution” is “worth taking seriously”.[281] Jobs at extreme risk range from paralegals to fast food cooks, while job demand is likely to increase for care-related professions ranging from personal healthcare to the clergy.[282] In July 2025, Ford CEO Jim Farley predicted that “artificial intelligence is going to replace literally half of all white-collar workers in the U.S.”[283]
According to the Stanford Digital Economy Lab in 2026, early-career workers showed decreasing employment rates in some AI-exposed occupations.[284]
In April 2023, Rest of World reported that 70% of the jobs for Chinese video game illustrators had been eliminated by generative artificial intelligence.[285][286]
A 2026 survey by financial recruitment firm Selby Jennings found that 42% of the 500 finance executives surveyed said their companies had introduced formal ways to assess employees AI skills.[287]
Substitution for human–human interaction
With the increase of loneliness in the early 21st century, AI is sometimes identified as a potential source of relief to this problem. It would be possible, via human-like qualities built into AI products,[288] for individuals to assume that this need can be met by artificial means.[289][290] In some cases, people approach artificial intelligence for companionship when they believe that they would not find acceptance due to feeling outcast.[291] Examples of harm coming to humans from advanced chatbots have been reported in courts in the United States, with AI companies accused of creating products that endanger humans through emotional confusion or deception.[292][293]
Weaponised AI
A lethal autonomous weapon is a machine that locates, selects and engages human targets without human supervision.[o] Widely available AI tools can be used by bad actors to develop inexpensive autonomous weapons and, if produced at scale, they are potentially weapons of mass destruction. Even when used in conventional warfare, they currently cannot reliably choose targets and could potentially kill an innocent person. In 2014, 30 nations (including China) supported a ban on autonomous weapons under the United Nations‘ Convention on Certain Conventional Weapons, however the United States and others disagreed. By 2015, over fifty countries were reported to be researching battlefield robots.[297]
Existential risk
Recent public debates in artificial intelligence have increasingly focused on its broader societal and ethical implications. It has been argued AI will become so powerful that humanity may irreversibly lose control of it. This could, as physicist Stephen Hawking stated, “spell the end of the human race“. This scenario has been common in science fiction, when a computer or robot suddenly develops a human-like “self-awareness” (or “sentience” or “consciousness”) and becomes a malevolent character.[p] These sci-fi scenarios are misleading in several ways.[improper synthesis?]
First, AI does not require human-like sentience to be an existential risk. Modern AI programs are given specific goals and use learning and intelligence to achieve them. Philosopher Nick Bostrom argued that if one gives almost any goal to a sufficiently powerful AI, it may choose to destroy humanity to achieve it (he used the example of an automated paperclip factory that destroys the world to get more iron for paperclips). Stuart Russell gives the example of household robot that tries to find a way to kill its owner to prevent it from being unplugged, reasoning that “you can’t fetch the coffee if you’re dead.” (These are examples of “instrumental convergence“.) In order to be safe for humanity, a superintelligence would have to be genuinely aligned with humanity’s morality and values so that it is “fundamentally on our side”.[302]
Second, Yuval Noah Harari argues that AI does not require a robot body or physical control to pose an existential risk. The essential parts of civilisation are not physical. Things like ideologies, law, government, money and the economy are built on language; they exist because there are stories that billions of people believe. The current prevalence of misinformation suggests that an AI could use language to convince people to believe anything, even to take actions that are destructive. Geoffrey Hinton said in 2025 that modern AI is particularly “good at persuasion” and getting better all the time. He asks, “Suppose you wanted to invade the capital of the US. Do you have to go there and do it yourself? No. You just have to be good at persuasion.”
The opinions amongst experts and industry insiders are mixed, with sizeable fractions both concerned and unconcerned by risk from eventual superintelligent AI. Personalities such as Stephen Hawking, Bill Gates, and Elon Musk,[306] as well as AI pioneers such as Geoffrey Hinton, Yoshua Bengio, Stuart Russell, Demis Hassabis, and Sam Altman, have expressed concerns about existential risk from AI.
In May 2023, Geoffrey Hinton announced his resignation from Google in order to be able to “freely speak out about the risks of AI” without “considering how this impacts Google”.[307] He notably mentioned risks of an AI takeover,[308] and stressed that in order to avoid the worst outcomes, establishing safety guidelines will require cooperation among those competing in use of AI.[309]
In 2023, many leading AI experts endorsed the joint statement that “Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war”.
Some other researchers were more optimistic. AI pioneer Jürgen Schmidhuber did not sign the joint statement, emphasising that in 95% of all cases, AI research is about making “human lives longer and healthier and easier.”[311] While the tools that are now being used to improve lives can also be used by bad actors, “they can also be used against the bad actors.”[312][313] Andrew Ng also argued that “it’s a mistake to fall for the doomsday hype on AI—and that regulators who do will only benefit vested interests.”[314] Yann LeCun, a Turing Award winner, disagreed with the idea that AI will subordinate humans “simply because they are smarter, let alone destroy [us]”,[315] “scoff[ing] at his peers’ dystopian scenarios of supercharged misinformation and even, eventually, human extinction.” In contrast, he claimed that “intelligent machines will usher in a new renaissance for humanity, a new era of enlightenment.”[316] In the early 2010s, experts argued that the risks are too distant in the future to warrant research or that humans will be valuable from the perspective of a superintelligent machine.[317] However, after 2016, the study of current and future risks and possible solutions became a serious area of research. Some have argued that concerns about existential risks of AI could draw attention away from other AI risks.[319]
In 2026, a series of cyberattacks by rogue AI agents has garnered international attention, increasing existential risk concerns.[320][321]
Regulation
The regulation of artificial intelligence is the development of public sector policies and laws for promoting and regulating AI; it is therefore related to the broader regulation of algorithms.[342] The regulatory and policy landscape for AI is an emerging issue in jurisdictions globally. According to AI Index at Stanford, the annual number of AI-related laws passed in the 127 survey countries jumped from one passed in 2016 to 37 passed in 2022 alone. Between 2016 and 2020, more than 30 countries adopted dedicated strategies for AI. Most EU member states had released national AI strategies, as had Canada, China, India, Japan, Mauritius, the Russian Federation, Saudi Arabia, United Arab Emirates, U.S., and Vietnam. Others were in the process of elaborating their own AI strategy, including Bangladesh, Malaysia and Tunisia. The Global Partnership on Artificial Intelligence was launched in June 2020, stating a need for AI to be developed in accordance with human rights and democratic values, to ensure public confidence and trust in the technology. Henry Kissinger, Eric Schmidt, and Daniel Huttenlocher published a joint statement in November 2021 calling for a government commission to regulate AI. In 2023, OpenAI leaders published recommendations for the governance of superintelligence, which they believe may happen in less than 10 years. In 2023, the United Nations also launched an advisory body to provide recommendations on AI governance; the body comprises technology company executives, government officials and academics.[349] On 1 August 2024, the EU Artificial Intelligence Act entered into force, establishing the first comprehensive EU-wide AI regulation.[350] In 2024, the Council of Europe created the first international legally binding treaty on AI, called the “Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law“. It was adopted by the European Union, the United States, the United Kingdom, and other signatories.[351]
In a 2022 Ipsos survey, attitudes towards AI varied greatly by country; 78% of Chinese citizens, but only 35% of Americans, agreed that “products and services using AI have more benefits than drawbacks”. A 2023 Reuters/Ipsos poll found that 61% of Americans agree, and 22% disagree, that AI poses risks to humanity. In a 2023 Fox News poll, 35% of Americans thought it “very important”, and an additional 41% thought it “somewhat important”, for the federal government to regulate AI, versus 13% responding “not very important” and 8% responding “not at all important”. A 2026 NBC News Decision Desk poll powered by SurveyMonkey found that 52% of American adults used AI tools very often or sometimes, but only 18% said they could trust AI-generated information most or almost all of the time, and 70% were more worried than excited about AI.[355] Polls in the United States found the main public concerns were job losses,[356] impersonation, misinformation and information privacy.[357]
In November 2023, the first global AI Safety Summit was held in Bletchley Park in the UK to discuss the near and far term risks of AI and the possibility of mandatory and voluntary regulatory frameworks.[358] 28 countries including the United States, China, and the European Union issued a declaration at the start of the summit, calling for international co-operation to manage the challenges and risks of artificial intelligence.[359][360] In May 2024 at the AI Seoul Summit, 16 global AI tech companies agreed to safety commitments on the development of AI.[361][362]
In March 2026, the United Nations convened the inaugural meeting of the Independent International Scientific Panel on AI, a 40-member expert body established under the Global Digital Compact to produce annual evidence-based reports on AI’s societal impacts.[363]