We develop a model of advertising markets in an environment where consumers may switch (or “multi-home”) across publishers. Consumer switching generates inefficiency in the process of matching ...advertisers to consumers, because advertisers may not reach some consumers and may impress others too many times. We find that when advertisers are heterogeneous in their valuations for reaching consumers, the switching-induced inefficiency leads lower-value advertisers to advertise on a limited set of publishers, reducing the effective demand for advertising and thus depressing prices. As the share of switching consumers expands (e.g., when consumers adopt the Internet for news or increase their use of aggregators), ad prices fall. We demonstrate that increased switching creates an incentive for publishers to invest in quality as well as extend the number of unique users, because larger publishers are favored by advertisers seeking broader “reach” (more unique users) while avoiding inefficient duplication.
This paper was accepted by Bruno Cassiman, business strategy
.
Background
Several challenging clinical situations in patients with peritonitis can result in an open abdomen (OA) and subsequent temporary abdominal closure (TAC). Indications and treatment choices ...differ among surgeons. The risk of fistula development and the possibility to achieve delayed fascial closure differ between techniques. The aim of this study was to review the literature on the OA and TAC in peritonitis patients, to analyze indications and to assess delayed fascial closure, enteroatmospheric fistula and mortality rate, overall and per TAC technique.
Methods
Electronic databases were searched for studies describing the OA in patients of whom 50 % or more had peritonitis of a non-traumatic origin.
Results
The search identified 74 studies describing 78 patient series, comprising 4,358 patients of which 3,461 (79 %) had peritonitis. The overall quality of the included studies was low and the indications for open abdominal management differed considerably. Negative pressure wound therapy (NPWT) was the most frequent described TAC technique (38 of 78 series). The highest weighted fascial closure rate was found in series describing NPWT with continuous mesh or suture mediated fascial traction (6 series, 463 patients: 73.1 %, 95 % confidence interval 63.3–81.0 %) and dynamic retention sutures (5 series, 77 patients: 73.6 %, 51.1–88.1 %). Weighted rates of fistula varied from 5.7 % after NPWT with fascial traction (2.2–14.1 %), 14.6 % (12.1–17.6 %) for NPWT only, and 17.2 % after mesh inlay (17.2–29.5 %).
Conclusion
Although the best results in terms of achieving delayed fascial closure and risk of enteroatmospheric fistula were shown for NPWT with continuous fascial traction, the overall quality of the available evidence was poor, and uniform recommendations cannot be made.
•New developments in AI are modelled as a reduction in the cost of prediction.•Predictions improve decisions but are imperfect.•There is a role for human judgment in adjusting decisions.•As the ...number of decisions rises, there is a diseconomy associated with the scope of decisions allocated to a human versus a machine.•This will impact on the construction of tasks as part of jobs.
Based on recent developments in the field of artificial intelligence (AI), we examine what type of human labor will be a substitute versus a complement to emerging technologies. We argue that these recent developments reduce the costs of providing a particular set of tasks – prediction tasks. Prediction about uncertain states of the world is an input into decision-making. We show that prediction allows riskier decisions to be taken and this is its impact on observed productivity although it could also increase the variance of outcomes as well. We consider the role of human judgment in decision-making as prediction technology improves. Judgment is exercised when the objective function for a particular set of decisions cannot be described (i.e., coded). However, we demonstrate that better prediction impacts the returns to different types of judgment in opposite ways. Hence, not all human judgment will be a complement to AI. Finally, we show that humans will delegate some decisions to machines even when the decision would be superior with human input.
When do scientists and other innovators organize into collaborative teams, and why do they do so for some projects and not others? At the core of this important organizational choice is, we argue, a ...trade-off scientists make between the productive efficiency of collaboration and the credit allocation that arises after the completion of collaborative work. In this paper, we explore this trade-off by developing a model to structure our understanding of the factors shaping researcher collaborative choices, in particular the implicit allocation of credit among participants in scientific projects. We then use the annual research activity of 661 faculty scientists at the Massachusetts Institute of Technology over a 31-year period to explore the trade-off between collaboration and reward at the individual faculty level and to infer critical parameters in the collaborative organization of scientific work.
This paper was accepted by Lee Fleming, entrepreneurship and innovation
.
Analyses of artificial intelligence (AI) adoption focus on its adoption at the individual task level. What has received significantly less attention is how AI adoption is shaped by the fact that ...organizations are composed of many interacting tasks. AI adoption may, therefore, require system‐wide change, which is both a constraint and an opportunity. We provide the first formal analysis where multiple tasks may be part of an interdependent system. We find that reliance on AI, a prediction tool, increases decision variation, which, in turn, raises challenges if decisions across the organization interact. Reducing inter‐dependencies between decisions softens that impact and can facilitate AI adoption. However, it does this at the expense of synergies. By contrast, when there are mechanisms for inter‐decision coordination, AI adoption is enhanced when there are more inter‐dependencies. Consequently, we show that there are important cases where AI adoption will be enhanced when it can be adopted beyond tasks but as part of a designed organizational system.
•It presents a dynamic model without assumed direct interactions between patenting and publishing.•The model shows when patents and publications are complements or substitutes.•The model shows that ...complementarity is driven by the similar costs associated with patent and publication disclosures and the need to use non-wage instruments to recruit scientists.•Stronger intellectual property rights can, therefore, lead to more openness in science even in commercial settings.•Races for priority can also lead to more disclosure in commercial science.
This paper provides a theoretical investigation of the tension over knowledge disclosure between firms and their scientific employees. While empirical research suggests that scientists exhibit a “taste for science,” such open disclosures can limit a firm's competitive advantage or ability to profitably commercialize their innovations. To explore how this tension is resolved we focus on the strategic interaction between researchers and firms bargaining over whether (and how) knowledge will be disclosed. We evaluate four disclosure strategies: secrecy, patenting, open science (scientific publication) and patent-paper pairs providing insights into the determinants of the disclosure strategy of a firm. We find that patents and publications can be complementary instruments facilitating the disclosure of knowledge-providing predictions as to when stronger IP protection regimes might drive openness by firms.
This paper presents a synthetic framework identifying the central drivers of start-up commercialization strategy and the implications of these drivers for industrial dynamics. We link strategy to the
...commercialization environment—the microeconomic and strategic conditions facing a firm that is translating an “idea” into a value proposition for customers. The framework addresses why technology entrepreneurs in some environments undermine established firms, while others cooperate with incumbents and reinforce existing market power. Our analysis suggests that competitive interaction between start-up innovators and established firms depends on the presence or absence of a “market for ideas”. By focusing on the operating requirements, efficiency, and institutions associated with markets for ideas, this framework holds several implications for the management of high-technology entrepreneurial firms.
Background
There is a clear association between hyperglycaemia and surgical‐site infection (SSI). Intensive glucose control may involve a risk of hypoglycaemia, which in turn results in potentially ...severe complications. A systematic review was undertaken of studies comparing intensive versus conventional glucose control protocols in relation to reduction of SSI and other outcomes, including hypoglycaemia, mortality and stroke.
Methods
PubMed, Embase, CENTRAL, CINAHL and WHO databases from 1 January 1990 to 1 August 2015 were searched. Inclusion criteria were RCTs comparing intensive with conventional glucose control protocols, and reporting on the incidence of SSI. Meta‐analyses were performed with a random‐effects model, and meta‐regression was subsequently undertaken. Targeted blood glucose levels, achieved blood glucose levels, and important adverse events were summarized.
Results
Fifteen RCTs were included. The summary estimate showed a significant benefit for an intensive compared with a conventional glucose control protocol in reducing SSI (odds ratio (OR) 0·43, 95 per cent c.i. 0·29 to 0·64; P < 0·001). A significantly higher risk of hypoglycaemic events was found for the intensive group compared with the conventional group (OR 5·55, 2·58 to 11·96), with no increased risk of death (OR 0·74, 0·45 to 1·23) or stroke (OR 1·37, 0·26 to 7·20). These results were consistent both in patients with and those without diabetes, and in studies with moderately strict and very strict glucose control.
Conclusion
Stricter and lower blood glucose target levels of less than 150 mg/dl (8·3 mmol/l), using an intensive protocol in the perioperative period, reduce SSI with an inherent risk of hypoglycaemic events but without a significant increase in serious adverse events.
Low glucose prevents surgical‐site infection
Artificial Intelligence Agrawal, Ajay; Gans, Joshua S.; Goldfarb, Avi
The Journal of economic perspectives,
04/2019, Letnik:
33, Številka:
2
Journal Article
Recenzirano
Odprti dostop
Recent advances in artificial intelligence are primarily driven by machine learning, a prediction technology. Prediction is useful because it is an input into decision-making. In order to appreciate ...the impact of artificial intelligence on jobs, it is important to understand the relative roles of prediction and decision tasks. We describe and provide examples of how artificial intelligence will affect labor, emphasizing differences between when the automation of prediction leads to automating decisions versus enhancing decision-making by humans.
PDF
