Similar to the crisis of accountability in criminal law, which involves an artificial intelligence system committing the actus reus of the crime in the absence of mens rea, the gig economy has itself emerged as a crisis of legality. In the wake of the global spread of the digital platform economy, the structure of the relationship between capital and labour has undergone a significant restructuring, heralding the emergence of a completely new paradigm in which hierarchical structures have ceased to exist in favour of algorithmic structures.[1] In essence, a complex algorithmic structure has been implemented in which the automated process acts not only as the dispatching body, but also as the overseer of the process and the executioner of the worker’s career prospects. Human managers have long been superseded by automated processes controlled through algorithms and other technologies that determine task delegation, dynamically determine the amount of pay, and conduct oversight with extreme precision. While the platform operators tout the effectiveness of their digital facilitation model, there is a hidden danger that affects millions of workers: “termination by algorithm,” also called app deactivation.
Indeed, the contemporary concept of the gig economy significantly transforms the conventional system of employment. The fact that platform workers tend to be categorised as independent contractors and not statutory employees means that they exist within a legal grey area where there are no typical forms of employment law, such as minimum wages, dismissal protection, and collective bargaining laws.[2] As such, digital labour platforms have almost complete autonomy in deciding whom to deactivate. Deactivation is defined as a process by which an algorithm, temporarily or permanently, removes access of a particular user to the digital platform, thus terminating his or her earning capacity.
Algorithms used in platform work are often portrayed by platform creators as a purely objective and impartial system of optimisation. Empirical analysis shows that algorithms work as a tool to discriminate against vulnerable groups. They focus on customer ratings, acceptance rate, and job completion score. Since these algorithms rely on inputs provided by users without filtering any bias, they are a mechanism to encode bias into an automated system. A telling example shows how the algorithm works to discriminate against people performing platform work. It is known that platforms reward constant availability and presence as positive qualities. In turn, an algorithm that promotes such behaviour would discriminate against those who can hardly stay available due to household commitments, which affects primarily women.[3] Second, algorithmic management lacks contextual awareness. Algorithms cannot distinguish between weather conditions and a late arrival of a courier. As a consequence, error rates of automated termination are incredibly high[4] since many workers get deactivated due to technical faults or customer complaints that show biasness.[5]
Arbitrary deactivation has emerged as a crucial constitutional battleground in India, one of the few nations in the world currently experiencing the rapid growth of the gig economy. Gig workers’ informal status and consequent exclusion from social protections have intensified their struggle at the constitutional level.[6] The most significant litigation challenge that has been launched against the concept of algorithmic management in the Global South concerns the case initiated in the Supreme Court of India – The Indian Federation of App-Based Transport Workers (IFAT) v. Union of India.[7] In the case, it has been argued by the plaintiffs that the operational models implemented by aggregators, and in particular, their discretion to deactivate an account, violate several articles of the Constitution of India, including Articles 14, 21, and 23. Building on the established precedent from the Olga Tellis v. Bombay Municipal Corporation case, it can be claimed that deactivation, for a gig worker, amounts to an unexplained, irrational termination of his/her means of living and therefore violates the right to life under Article 21.
However, while the Code on Social Security passed by the Indian Parliament took a purely “welfare-centric” approach to avoid addressing the basic issues underlying the power structure of algorithmic management, several sub-national jurisdictions have made groundbreaking moves to hold algorithms accountable. The government of the State of Karnataka, for instance, made significant strides towards regulating the issue with comprehensive laws that openly challenge information asymmetry in platform capitalism. Under Karnataka’s provisions, an aggregator cannot terminate or deactivate a gig worker without justifiable reasons stated in writing accompanied by a notice period of fourteen days in line with the principles of natural justice. Furthermore, each registered platform is expected to set up an Internal Dispute Resolution Committee (IDRC), which integrates seamlessly with Karnataka’s public grievance portal.[8]
On the other hand, the European approach to algorithmic control has been much more confrontational by using the well-developed frameworks of data protection and privacy laws. The GDPR has spearheaded the struggle for gig workers’ rights, particularly through Article 22, which entitles individuals the right not to be subjected to any decision based on automated processing alone. This particular clause has been scrutinised thoroughly in the context of an ongoing lawsuit brought by ride-hailing drivers against sudden suspensions by the District Court of Amsterdam and the Dutch Court of Appeal.[9] According to the platform’s argument, human employees would review the relevant information in the file prior to the ultimate decision of firing; thus, they were not in violation of Article 22. The Court has explicitly dismissed this argument by emphasising the need for substantial human involvement to sidestep the ban.
With this in mind, the European Union developed the Platform Work Directive, which states that decisions made by artificial intelligence that may affect terms and conditions of employment must be subject to review by employees with the legal competence to override the decision of the automated system.[10] Alongside this data-oriented approach, the United Kingdom took on statutory employment classification. In the landmark case of Uber BV v. Aslam et al., the Supreme Court tackled the issue of power asymmetry by repudiating the doctrine of “contractual supremacy.” Statutory benefits cannot be stripped away with clever legal maneuvers, according to the court; the extreme level of managerial control represented by algorithmic management, including deactivation, is enough to classify drivers as “workers.”[11]
The U.S. gig economy functions within an “at-will” employment system where platforms strongly defend classifying workers as independent contractors. Measures like California’s Proposition 22 offer limited protections against account deactivation but contain major loopholes, allowing companies to bypass fairness requirements under broad claims of fraud or safety concerns.[12]
It is clear from above that the deactivation of algorithms poses risks for gig workers in terms of their livelihoods, respect, and procedural justice. The evolution of legal norms in India, the European Union, the UK, and the USA is a testament to the understanding that decisions regarding employment should not depend solely on automation. Although the focus of the EU and the UK has been on transparency and human involvement, even in the USA, such protections are being demanded by local jurisdictions.
Author: Priyanka Ratha (Vikash Law School, Bargarh)
References:
[1] Tim Green, ‘The Boss in the Code: Labour Rights When Machines Manage You’ (SmarterArticles, 11 May 2026) https://smarterarticles.co.uk/the-boss-in-the-code-labour-rights-when-machines-manage-you accessed 29 May 2026.
[2] Economic Policy Institute, ‘Flexible Work without Exploitation: Reversing Tech Companies’ State-by-State Agenda to Unravel Workers’ Rights and Misclassify Workers as “Contractors” in the Gig Economy and Beyond’ (Economic Policy Institute, 10 June 2024) https://www.epi.org/publication/flexible-work-without-exploitation/ accessed 29 May 2026.
[3] International Labour Organization, The Role of Digital Labour Platforms in Transforming the World of Work (International Labour Office 2021) https://www.ilo.org/global/research/global-reports/weso/2021/WCMS_771749/lang–en/index.htm accessed 29 May 2026.
[4] Valerio De Stefano, ‘The Rise of the “Just-in-Time Workforce”: On-Demand Work, Crowd Work and Labour Protection in the “Gig-Economy”’ (2015) Comparative Labor Law & Policy Journal (forthcoming) https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2682602 accessed 29 May 2026.
[5] Human Rights Watch, Algorithms of Exploitation: Rights Abuses in the Gig Economy and the Global Fight for Change (Human Rights Watch, 13 May 2026) https://www.hrw.org/feature/2026/05/13/algorithms-of-exploitation/rights-abuses-in-the-gig-economy-and-the-global-fight accessed 29 May 2026.
[6] The Lawscape, ‘Gig Worker Protection in India: The Constitutional Gap in the Platform Economy’ (The Lawscape, 2026) https://www.thelawscape.in/gig-worker-protection-constitutional-gap-india/ accessed 29 May 2026.
[7] Supreme Court Observer, ‘Gig Workers’ Access to Social Security: Writ Petition Summary (The Indian Federation of App-based Transport Workers v Union of India)’ (Supreme Court Observer, 7 December 2021) https://www.scobserver.in/reports/gig-workers-access-to-social-security-the-indian-federation-of-app-based-transport-workers-ifat-v-union-of-india-writ-petition-summary/ accessed 29 May 2026.
[8] Chambers and Partners, ‘Digitizing Justice: Grievance Redressal for Platform-Based Gig Workers in Karnataka’ (Chambers and Partners, 18 May 2026) https://chambers.com/articles/digitizing-justice-grievance-redressal-for-platform-based-gig-workers-in-karnataka accessed 29 May 2026.
[9] Guillermo Lazcoz, ‘Automated Decision-Making Under Amsterdam’s District Court Judgements: Drivers v Uber and Ola’ in Data Protection and the New Digital Workplace (Larcier 2021) https://www.researchgate.net/publication/356531942_Automated_Decision-Making_Under_Amsterdam%27s_District_Court_Judgements_Drivers_v_Uber_and_Ola accessed 29 May 2026.
[10] Arthur Cox LLP, ‘A Deep Dive into the EU Platform Work Directive’ (Arthur Cox, 16 February 2026) https://www.arthurcox.com/knowledge/a-deep-dive-into-the-eu-platform-work-directive/ accessed 29 May 2026.
[11] Record of Law, ‘Uber BV and Others v Aslam and Others [2021] UKSC 5’ (Record of Law, 3 March 2026) https://recordoflaw.in/uber-bv-and-others-v-aslam-and-others-2021-uksc-5/ accessed 29 May 2026.
[12] CalMatters, ‘Rideshare Drivers Sue Uber over Being Kicked off App in New Challenge to California Law’ (CalMatters, 20 April 2026) https://calmatters.org/economy/2026/04/uber-proposition22-lawsuit/ accessed 29 May 2026.
