Computer vision, a core domain of artificial intelligence (AI), is the field that enables the computational analysis, understanding, and generation of visual data. Despite being historically rooted in military funding and increasingly deployed in warfare, the field tends to position itself as a neutral, purely technical endeavor, failing to engage in discussions about its dual-use applications. Yet it has been reported that computer vision systems are being systematically weaponized to assist in technologies that inflict harm, such as surveillance or warfare. Expanding on these concerns, we study the extent to which computer vision research is being used in the military and surveillance domains. We do so by collecting a dataset of tech companies with financial ties to the field's central research exchange platform: conferences. Conference sponsorship, we argue, not only serves as strong evidence of a company's investment in the field but also provides a privileged position for shaping its trajectory. By investigating sponsors' activities, we reveal that 44% of them have a direct connection with military or surveillance applications. We extend our analysis through two case studies in which we discuss the opportunities and limitations of sponsorship as a means for uncovering technological weaponization.
Figures & tables
Figure 1. The CVWeap dataset collection process consists of four parts: computer vision sponsors, metadata, domain of weaponization, and involvement profile.
editions
editions
sponsors
sponsors
sponsors
conference
since
location
2004-2024
CVWeap
entries
unique
CVWeap
CVPR
1983
United States
21
17
929
342
334
ICCV
1987
International
10
9
228
162
162
ECCV
1990
Europe
11
8
245
178
149
overall
42
36
1,402
503
469
Table 1. Details on computer vision conferences and sponsorship in the CVWeap dataset.
name
scope
size
domain
refs
AI War Cloud Database ( Ciston, 2025 )
AI weapons and companies developing them.
40
military
✓
AIGS Index ( Feldstein, 2019 )
AI and big data surveillance use for 179 countries and companies involved.
99
surveillance
✓
Atlas of Surveillance ( Maass et al., 2025 )
Surveillance technologies deployed by law enforcement across the US.
342
surveillance
✓
Investigate ( AFSC, 2025 )
Companies involved in Israeli military occupation.
255
military
✓
SIPRI Arms Industry Database ( SIPRI, 2024 )
Arms-producing and military services companies.
256
military
-
Surveillance Watch ( Surveillance Watch, 2025 )
Companies that develop technologies to enable tracking individuals.
684
surveillance
✓
Table 2. External databases used to cross-reference sponsors. Size indicates number of companies in the database, and refs whether the database contains source references.
Figure 2. Domain of application and involvement profile annotation flowchart.
Figure 3. Domain of application and involvement profile statistics in the CVWeap dataset. Left: percentage of sponsors per domain of application, with each square representing 1% of the sponsors. Middle: involvement profile (overt, transparent, or opaque) of the sponsors in the military domain. Right: involvement profile of the sponsors in the surveillance domain.
Figure 4. Number of sponsors per edition in each of the three main computer vision conferences.
Figure 5. Where are computer vision sponsors from? Number of sponsors per country in the CVWeap dataset.
Figure 6. Number of sponsors per country with at least a sponsor involved in the military or surveillance domain. Left: number of sponsors per country and domain of application, sorted in descending order by the number of military + surveillance sponsors. Right: number of sponsors in the military + surveillance domain normalized by the country’s GDP.
Figure 7. Percentage of sponsors in the surveillance and military domain per computer vision subfield.
Computer vision saliency models predict where people will look, one map per image, and a predicted-attention industry sells those maps in place of measuring real viewers. I evaluate the leading models against every publicly available eye-tracking corpus that records viewer demographics, comprising 27,384 participants across eight datasets, including 11.4 million webcam gaze points from 2,993 US adults recruited to national quotas while viewing circulating news images. I show that a single map predicts where every group looks about equally well, but systematically ignores how long each place holds a group's attention. That time follows a clear sociodemographic and political fingerprint, so the map carries a systematic bias against the youngest and oldest viewers and against Black, lower-income, and Republican-leaning ones. I use each group's own gaze to test whether it carries a learnable pattern, across every demographic axis the sample supports. Showing which audiences these systems can learn and what determines this sets the standard for judging claims that computer vision captures diverse human vision.
Artificial intelligence (AI) technologies are increasingly used in modern weapons systems. Notably, these systems have recently been involved in mass killings and destruction at scale. Furthermore, there is currently a strong interest and competition among powerful players to accelerate the proliferation of weapons with automated or AI-based components, a phenomenon known as AI arms race. This competition poses a risk of causing even more deaths and devastation in the future, as well as increased power and wealth inequality. In this work, we aim to shed light on the role of AI researchers as implicated subjects in the harms caused by weapons enabled by AI technologies. We investigate and discuss the specifics of this implication and explore ways to transfigure this position of implication into one of differentiated, long-distance solidarity with the victims of technologically fortified injustices.
Alexandra Volokhova, Alex Hernandez-Garcia
Universit´e de Montr´eal, Mila – Quebec AI Institute Montreal, QC, Canada
While conventional camera-based surveillance systems and metal detectors can provide a sufficient level of security when used appropriately, they have fundamental limitations that reduce their applicability. Camera-based systems are unable to detect occluded or concealed threats and metal detectors require each person to enter through a narrow opening one at a time, leading to congestion issues and limited use in large crowds. A WiFi-based security system has the potential to be a complementary modality that is able to handle occlusions or concealment and is not hindered by strict spatial configurations. In this paper, we propose use Channel State Information estimated from WiFi-signals to detect concealed handguns. To this end we collect a dataset of aligned WiFi signals and RGB images that we call the WiFire dataset. We also propose a deep learning model called WiFireNet tailored to detect concealed handguns in WiFi signals using vision-based supervision signals. Our results show that concealed handgun detection is possible in WiFi and has promise as a complementary modality to surveillance cameras and metal detectors. We release the dataset, model and code to facilitate the further research into WiFi-based handgun detection.
Erik Persson, David Gustafsson, Tomas Wilkinson
Swedish Defence Research Agency, FOI, SE-164 90 Stockholm, Sweden